Greenhouse gas emissions and fossil energy demand from ... · demand from small ruminant supply...
Transcript of Greenhouse gas emissions and fossil energy demand from ... · demand from small ruminant supply...
Greenhouse gas emissions and fossil energy demand from small ruminant supply chains
Guidelines for quanti�cation
DRAFT FOR PUBLIC REVIEW
Greenhouse gas emissions and fossil energy demand from small ruminant supply chains
Guidelines for quanti�cation
DRAFT FOR PUBLIC REVIEW
The designations employed and the presentation of material in this information product do not imply the expression of any opinion whatsoever on the part of the Food and Agriculture Organization of the United Nations (FAO) concerning the legal or development status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The mention of specific companies or products of manufacturers, whether or not these have been patented, does not imply that these have been endorsed or recommended by FAO in preference to others of a similar nature that are not mentioned.
The views expressed in this information product are those of the author(s) and do not necessarily reflect the views of FAO.
All rights reserved. FAO encourages reproduction and dissemination of material in this information product. Non-commercial uses will be authorized free of charge, upon request. Reproduction for resale or other commercial purposes, including educational purposes, may incur fees. Applications for permission to reproduce or disseminate FAO copyright materials, and all queries concerning rights and licences, should be addressed by e-mail to [email protected] or to the Chief, Publishing Policy and Support Branch, Office of Knowledge Exchange, Research and Extension, FAO, Viale delle Terme di Caracalla, 00153 Rome, Italy.
© FAO 2014
iii
Table of Contents 1
2
FOREWORD ....................................................................................................................................... vii 3
ACKNOWLEDGEMENTS ................................................................................................................... ix 4
GLOSSARY ......................................................................................................................................... xii 5
PART 1: OVERVIEW AND GENERAL PRINCIPLES ................................................................... 1 6
1 INTENDED USERS AND OBJECTIVES ..................................................................................... 2 7
2 SCOPE ............................................................................................................................................ 2 8
2.1 Environmental impact categories addressed in the guidelines ................................................ 2 9
2.2 Application .............................................................................................................................. 3 10
3 STRUCTURE AND CONVENTIONS ........................................................................................... 4 11
3.1 Structure .................................................................................................................................. 4 12
3.2 Presentational conventions ...................................................................................................... 6 13
4 ESSENTIAL BACKGROUND INFORMATION AND PRINCIPLES ......................................... 6 14
4.1 A brief introduction to LCA .................................................................................................... 6 15
4.2 Environmental impact categories ............................................................................................ 7 16
4.3 Normative references .............................................................................................................. 8 17
4.4 Guiding principles ................................................................................................................. 10 18
5 LEAP AND THE PREPARATION PROCESS ............................................................................ 11 19
5.1 Development of sector-specific guidelines ........................................................................... 12 20
5.2 The Small ruminants TAG and the preparation process ........................................................ 13 21
5.3 Period of validity ................................................................................................................... 14 22
6 BACKGROUND INFORMATION ON SMALL RUMINANT SUPPLY CHAINS ................. 14 23
6.1 Background and context ........................................................................................................ 14 24
6.2 Diversity of small ruminant production systems ................................................................... 15 25
6.3 Diversity of small ruminant value chains .............................................................................. 17 26
6.4 Multi-functionality of small ruminant supply chains ............................................................ 19 27
6.5 Overview of global emissions from small ruminants ............................................................ 20 28
iv
PART 2: METHODOLOGY FOR QUANTIFICATION OF GREENHOUSE GAS EMISSIONS AND 1
FOSSIL ENERGY DEMAND OF SMALL RUMINANT PRODUCTS ........................................................... 21 2
7 DEFINITION OF PRODUCTS AND PRODUCTION SYSTEMS ............................................. 22 3
7.1 Products description .............................................................................................................. 22 4
7.2 Life cycle stages: modularity ................................................................................................. 22 5
8 GOAL AND SCOPE DEFINITION ............................................................................................. 23 6
8.1 Goal of the LCA study .......................................................................................................... 23 7
8.2 Scope of the LCA study ........................................................................................................ 24 8
8.3 Functional unit ....................................................................................................................... 24 9
8.4 System boundary ................................................................................................................... 26 10
8.4.1 General/scoping analysis ............................................................................................... 26 11
8.4.2 Criteria for system boundary ......................................................................................... 28 12
8.4.3 Material contribution and threshold .............................................................................. 29 13
8.4.4 Time boundary for data ................................................................................................. 30 14
8.4.5 Capital goods ................................................................................................................. 30 15
8.4.6 Ancillary activities ........................................................................................................ 30 16
8.4.7 Delayed emissions ......................................................................................................... 30 17
8.4.8 Carbon offsets ................................................................................................................ 30 18
8.5 Impact categories ................................................................................................................... 31 19
9 MULTI-FUNCTIONAL PROCESSES AND ALLOCATION .................................................... 31 20
9.1 General principles .................................................................................................................. 31 21
9.2 A decision tree to guide methodology choices ...................................................................... 32 22
9.2.1 Allocation of transport ................................................................................................... 37 23
9.3 Application of general principles for small ruminant systems and processes ....................... 38 24
10 COMPILING AND RECORDING INVENTORY DATA .......................................................... 40 25
10.1 General principles .................................................................................................................. 40 26
10.2 Requirements and guidance for the collection of data .......................................................... 42 27
10.2.1 Requirements and guidance for the collection of primary data ..................................... 43 28
10.2.2 Requirements and guidance for the collection and use of secondary data .................... 44 29
10.2.3 Approaches for addressing data gaps in LCI ................................................................. 45 30
10.3 Data quality assessment ......................................................................................................... 46 31
10.3.1 Data quality rules ........................................................................................................... 46 32
v
10.3.2 Data quality indicators ................................................................................................... 47 1
10.4 Uncertainty analysis and related data collection ................................................................... 47 2
10.4.1 Secondary activity data.................................................................................................. 48 3
10.4.2 Default/proxy data ......................................................................................................... 48 4
10.4.3 Inter- and intra-annual variability in emissions ............................................................. 48 5
10.4.4 Inter- and intra-annual variability in emissions ............................................................. 48 6
11 LIFE CYCLE INVENTORY ........................................................................................................ 49 7
11.1 Overview ............................................................................................................................... 49 8
11.2 Cradle-to-farm-gate ............................................................................................................... 49 9
11.2.1 Feed assessment ............................................................................................................ 51 10
11.2.2 Animal population and productivity .............................................................................. 53 11
11.2.3 Manure production and management ............................................................................ 59 12
11.2.4 Emissions from other farm-related inputs ..................................................................... 62 13
11.2.5 Multi-functional processes and allocation of GHG emissions between co-products .... 63 14
11.3 Transportation ....................................................................................................................... 68 15
11.4 Inclusion and treatment of land-use-change .......................................................................... 69 16
11.5 Biogenic and soil carbon sequestration ................................................................................. 69 17
11.6 Primary processing stage ....................................................................................................... 69 18
11.6.1 Milk processing ............................................................................................................. 71 19
11.6.2 Fibre processing ............................................................................................................ 72 20
11.6.3 Meat processing ............................................................................................................. 74 21
11.6.4 On-site energy generation.............................................................................................. 77 22
12 INTERPRETATION OF LCA RESULTS .................................................................................... 77 23
12.1 Identification of key issues .................................................................................................... 78 24
12.2 Characterizing uncertainty ..................................................................................................... 78 25
12.2.1 Monte Carlo Analysis .................................................................................................... 79 26
12.2.2 Sensitivity analysis ........................................................................................................ 80 27
12.2.3 Normalization ................................................................................................................ 80 28
12.3 Conclusions, Recommendations and Limitations ................................................................. 80 29
12.4 Use and comparability of results ........................................................................................... 81 30
12.5 Good practice in reporting LCA results ................................................................................ 81 31
12.6 Report elements and structure ............................................................................................... 82 32
vi
12.7 Critical review ....................................................................................................................... 83 1
13 REFERENCES .............................................................................................................................. 84 2
APPENDICES ..................................................................................................................................... 90 3
Appendix 1: Greenhouse gas emissions from small ruminants: A review of existing methodologies 4
and guidelines ........................................................................................................................................ 91 5
Appendix 2: Small ruminants: Main producing countries .................................................................... 97 6
Appendix 3: Summary of carcass-weight: live-weight ratios (as percentages) for goats and sheep for 7
different regions .................................................................................................................................... 99 8
Appendix 4: Allocation options for manure exported off-farm .......................................................... 100 9
Appendix 5: Average sheep flock and goat herd parameters for different regions of the world ......... 102 10
Appendix 6: Calculation of enteric methane emissions from animal energy requirements ................ 103 11
12
13
vii
FOREWORD 1
The methodology developed in these draft guidelines aims to introduce a harmonized international 2
approach to the assessment of the environmental performance of small ruminant supply chains in a 3
manner that takes account of the specificity of the various production systems involved. It aims to 4
increase understanding of small ruminant supply chains and to help improve their environmental 5
performance. The guidelines are a product of the Livestock Environmental Assessment and 6
Performance (LEAP) Partnership, a multi-stakeholder initiative whose goal is to improve the 7
environmental sustainability of the livestock sector through better metrics and data. 8
The small ruminant1 sector is of worldwide importance. It comprises a wide diversity of systems that 9
provide a variety of products and functions. In 2011, sheep and goats produced more than 5 million 10
tonnes of meat and 24 million tonnes of milk. Production has increased by 1.7 percent and 1.3 percent 11
per year, respectively, during the past 20 years (FAOSTAT, 2013). This increase was driven mainly by 12
developing countries in Africa and Asia, however Oceania (mainly for meat) and Europe still contribute 13
significantly to production. Production systems can vary from intensive systems, wherein animals are 14
partially or predominantly housed, to extensive systems which rely on grazing and native forages, and 15
transhumance systems that involve large flock movements. Products are not restricted to meat and milk; 16
sheep are also valued for their wool (more than 2 million tonnes of greasy wool was produced in 2011), 17
and goats for their mohair and cashmere. Small ruminants also play a crucial role in sustaining 18
livelihoods in traditional, small-scale, rural and family-based production systems. Across the small 19
ruminant sector, there is strong interest in measuring and improving environmental performance. 20
In the development of these draft guidelines, the following objectives were regarded as key: 21
To develop a harmonized, science-based approach resting on a consensus among the sector’s 22
stakeholders; 23
To recommend a scientific but at the same time practical approach that builds on existing or 24
developing methodologies; 25
To promote an approach to assessment suitable for a wide range of small ruminant supply 26
chains; 27
To identify the principal areas where ambiguity or differing views exist as to the right 28
approach. 29
1 Small ruminants include goats, sheep, cervids and new world camelids (llamas and alpacas). These guidelines
focus on goats and sheep. Potential application to other small ruminant species is discussed in Section 2.2 and
10.2.3.
viii
Over the coming months these guidelines will be submitted to public review.2 The purpose will be to 1
strengthen the advice provided and ensure it meets the needs of those seeking to improve performance 2
through sound assessment practice. Nor is the present document intended to remain static. It will be 3
updated and improved as the sector evolves and more stakeholders become involved in LEAP, and as 4
new methodological frameworks and data become available. The development and inclusion of 5
guidance on the evaluation of additional environmental impacts is also viewed as a critical next step. 6
The strength of the guidelines developed within the LEAP Partnership across the various livestock 7
subsectors stems from the fact that they represent a coordinated cross-sectoral and international effort 8
to harmonize measurement approaches. Ideally, harmonization will lead to greater understanding, 9
transparent application and communication of metrics, and, importantly for the sector, real and 10
measurable improvement in performance. 11
12
13
Lalji Desai 14
LEAP Chair 15
February 2014 16
2 The public review period starts on 15 March 2014 and ends on 31 July 2014.
ix
ACKNOWLEDGEMENTS 1
These guidelines are a product of the Livestock Environmental Assessment and Performance (LEAP) 2
Partnership. Three groups contributed to their development: 3
The Technical Advisory Group (TAG) on small ruminants conducted the background research and 4
developed the core technical content of the guidelines. The small ruminants TAG was composed of: 5
Stewart Ledgard (TAG leader, AgResearch, New Zealand), Beverley Henry (Queensland University 6
of Technology, Australia), Marc Benoit (French National Institute for Agricultural Research, 7
France), C. Devendra (Independent Consultant, Malaysia), Jean-Baptiste Dollé and Armelle Gac 8
(Institut de l’Elevage, France), Christopher Lloyd (Organisation for the English beef and sheep 9
industry, EBLEX, UK) and Hans-Peter Zerfas (World Vision, Germany). 10
The LEAP Secretariat coordinated and facilitated the work of the TAG, guided and contributed to the 11
content development and ensured coherence between the various guidelines. The LEAP secretariat, 12
hosted at FAO, was composed of: Pierre Gerber (Coordinator), Alison Watson (Manager), Carolyn 13
Opio (Technical officer), Félix Teillard (Technical officer) and Aimable Uwizeye (Technical officer). 14
Laura Drauker (World Resource Institute), John Kazer (Carbon Trust) and Christel Cederberg (SIK 15
and Chalmers University of Technology, Gothenburg) assisted the Secretariat in reviewing the 16
guidelines. 17
The LEAP Steering Committee provided overall guidance for the activities of the Partnership and 18
helped review and cleared the guidelines for public release. During development of the guidelines the 19
LEAP Steering Committee was composed of: 20
Steering committee members 21
Douglas Brown (World Vision) 22
Elsa Delcombel (Government of France) 23
Lalji Desai (World Alliance of Mobile Indigenous People) – Chair 2013 to 2014 24
Jan Grenz (Government of Switzerland) 25
Vincent Guyonnet (International Egg Commission / International Poultry Council) 26
Dave Harrison (International Meat Secretariat) 27
Hsin Huang (International Meat Secretariat) 28
Giuseppe Luca Capodieci (The European Livestock And Meat Trading Union) 29
Delanie Kellon (International Dairy Federation) 30
Lionel Launois (Government of France) 31
Pablo Manzano (International Union for Conservation of Nature) 32
Nicolas Martin (European Feed Manufacturers’ Federation) 33
Paul Melville (Government of New Zealand) 34
x
Paul McKiernan (Government of Ireland) 1
Frank Mitloehner (University of California, Davis) – Chair 2012 to 2013 2
Anne-Marie Neeteson-van Nieuwenhoven (International Poultry Council) 3
Frank O’Mara (Irish Agriculture and Food Development Authority, Teagasc) 4
Antonio Onorati (International Planning Committee for World Food Sovereignty) 5
Lara Sanfrancesco (International Poultry Council) 6
Fritz Schneider (Bern University) 7
Rogier Schulte (Government of Ireland) 8
Henning Steinfeld (Food and Agriculture Organization) 9
Bryan Weech (World Wildlife Fund) 10
Geert Westenbrink (Government of the Netherlands) 11
Hans-Peter Zerfas (World Vision) 12
Observers 13
Rudolph De Jong (International Wool Textile Organization) 14
Matthias Finkbeiner (International Organization of Standards) 15
Michele Galatola (European Commission) 16
Sonia Valdivia (United Nations Environment Programme) 17
Elisabeth van Delden (International Wool Textile Organization) 18
LEAP is funded by its Members, with additional support from FAO and the Mitigation of Climate 19
Change in Agriculture (MICCA) Programme. 20
Although not directly responsible for the preparation of these guidelines, the TAGs on feed and 21
poultry indirectly contributed to their development through continuous exchanges during the 22
preparation phase. 23
xi
Recommended citation: LEAP, 2014. Greenhouse gas emissions and fossil energy demand from 1
small ruminant supply chains: Guidelines for quantification. Livestock Environmental Assessment 2
and Performance Partnership. FAO, Rome, Italy. 3
xii
GLOSSARY 1
Abbatoir An animal slaughterhouse.
Acidification Impact category that addresses impacts due to acidifying substances in the environment. Emissions of NOx, NH3 and SOx lead to releases of hydrogen ions (H+) when the gases are mineralized. The protons contribute to the acidification of soils and water when they are released in areas where the buffering capacity is low, resulting in forest decline and lake acidification.
Allocation Partitioning the input or output flows of a process between the product system under study and one or more different product systems.
Attributional Attributional Life Cycle Assessments focus on describing the environmentally relevant physical flows to and from a product or process.
Background process Stages of the supply chain which provide goods and services to the foreground system; not under the control of the study commissioner. See also: Foreground process.
Biogenic Derived from biomass, not from fossil sources.
Biomass Material of biological origin, excluding fossilized material or material embedded in geological formations.
Browse A general term applied to shrubs or trees that are fed on by goats by picking mouthfuls as they move.
Boundary Set of criteria specifying which unit processes are part of a product system (life cycle).
By-product Material produced during the processing (including slaughtering) of a livestock or crop product that is not the primary objective of the production activity (e.g. oil cakes, brans, offal or skins).
Capital goods Goods, such as machinery, equipment and buildings, used in the life cycle of products.
Carbon dioxide equivalent (CO2e)
Unit for comparing the radiative forcing (global warming potential) of a greenhouse gas expressed in terms of the amount of carbon dioxide that would have an equivalent impact.
Carbon footprint The level of greenhouse gas emissions produced by a particular activity or entity or product.
Carbon storage Retaining carbon of biogenic or atmospheric origin in a form other than atmospheric gas.
Carcass The body after slaughter from which the viscera, skin and head, and some other parts have been removed.
Cashmere Fine fibre from the Cashmere goat
Cold chain Refers to a system for distributing products in which the goods are constantly maintained at low temperatures (e.g. cold or frozen storage and transport), as they move from producer to consumer.
Combined heat and power (CHP)
Simultaneous generation in one process of useable thermal energy together with electrical and/or mechanical energy.
xiii
Compound feed/concentrate
Mixtures of feed materials which may contain additives for use as animal feed in the form of complete or complementary feedstuffs.
Consequential LCA Consequential LCA assessments describe how relevant environmental flows will change in response to different decisions.
Containers and packaging
Containers and packaging that reach consumers.
Co-production A multifunctional process that produces various outputs such as meat and milk. Production of the different goods cannot be varied, or only varied within a very narrow range.
Co-products Output from a production activity that generates more than one product (e.g. meat, milk, and, under some circumstances, litter are among the co-products of small ruminant production). The term does not include any services that may also be provided.
Cradle-to-gate Life cycle stages from the extraction or acquisition of raw materials to the point at which the product leaves the facility
Cull To reduce the size of a herd or flock by selling or killing a proportion of its members.
Data quality Characteristics of data that relate to their ability to satisfy stated requirements.
Direct energy Energy used on farms for livestock production activities (e.g. lighting, heating).
Doe Mature female goat.
Downstream emissions GHG emissions associated with processes that occur in the life cycle of a product subsequent to the processes owned or operated by the organization implementing these guidelines.
Emission factor Amount of greenhouse gases emitted, expressed as carbon dioxide equivalent and relative to a unit of activity (e.g. kg CO2e per unit input). NOTE: Emission factor data is obtained from secondary data sources.
Emissions Release to air or discharge to land and water that results in greenhouse gases entering the atmosphere. The main GHG emissions from agriculture are carbon dioxide (CO2), nitrous oxide (N2O) and methane (CH4).
Eutrophication
Nutrients (mainly nitrogen and phosphorus) from sewage outfalls and fertilized farmland accelerate the growth of algae and other vegetation in water. The degradation of organic material consumes oxygen resulting in oxygen deficiency and, in some cases, fish death. Eutrophication translates the quantity of substances emitted into a common measure expressed as the oxygen required for the degradation of dead biomass.
Ewe Mature female sheep usually over 2 years of age.
Feed conversion ratio Measure of the efficiency with which an animal converts feed into tissue, usually expressed in terms of kg of feed per kg of output (e.g. live weight or protein).
Feed digestibility Determines the relative amount of ingested feed that is actually absorbed by an animal and therefore the availability of feed energy or nutrients for growth, reproduction, etc.
Forage crop Crops, annual or biennial, grown to be used for grazing or harvested as a whole crop.
xiv
Foreground process The stages of the supply chain under the direct control of the commissioner of the LCA. For product developers and process operators, the foreground data is of special interest because direct changes in the system (changes in materials, designs, processes) have direct effects on the result, while the background system impacts can be influenced only indirectly (by the choices above). See also: Background process.
Functional unit Quantified performance of a product for use as a reference unit. It is essential that the functional unit allows comparisons that are valid where the compared objects (or time series data on the same object, for benchmarking) are comparable.
Global Warming Potential (GWP)
The intensity of a GHG’s warming potential, which is different for each gas. The factors used to convert GHGs into CO2 equivalents are defined by IPCC guidelines and can be found in Section 4.6.
Graze To feed directly on growing grass, pasturage or forage crops.
Greasy fibre Untreated fibre straight off an animal (e.g. raw wool, cashmere or mohair).
Greenhouse gases (GHGs)
Gaseous constituents of the atmosphere, both natural and anthropogenic, that absorb and emit radiation at specific wavelengths within the spectrum of infrared radiation emitted by the Earth’s surface, the atmosphere, and clouds. GHGs include carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), hydrofluoro-carbons (HFCs), perfluorocarbons (PFCs) and sulphur hexafluoride (SF6).
Hogget Young sheep between a lamb and an adult sheep (a two-tooth from approximately 10–16 months of age).
Human toxicity cancer
Impact category that accounts for the adverse health effects on human beings caused by the intake of toxic substances through inhalation of air, food/water ingestion and penetration through the skin, insofar as they are related to cancer.
Impact category A class representing environmental issues of concern to which life cycle inventory analysis results may be assigned.
Impact category indicator
Quantifiable representation of the contribution of a product unit to the specific impact.
Infrastructure Product not intended for consumption, with a lifetime exceeding one year. Synonym: Capital goods.
Input Product, material or energy flow that enters a unit process.
Joint production A multifunctional process in which production of the various outputs can be varied separately.
Kid Young male or female goat.
Lamb A young sheep from birth up until it is classified as a hogget, at approximately 12 months of age, although there is no specific age or time for this change.
Lanolin Also called wool fat. A yellowish viscous substance extracted from wool, consisting of a mixture of esters of fatty acids; used in some ointments.
LCA See Life Cycle Assessment.
Life cycle assessment Compilation and evaluation of inputs, outputs and the potential environmental impacts of a product system throughout its life cycle.
xv
Life cycle GHG emissions
Sum of GHG emissions resulting from all stages of the life cycle of a product and within the specified system boundaries of the product.
Life cycle Consecutive and interlinked stages of a product system, from raw material acquisition or generation of natural resources to end of life, inclusive of any recycling or recovery activity.
LCI See Life Cycle Inventory
Life cycle inventory Compilation of the exchanges for a unit process. These include purchased input materials, materials extracted from nature (e.g., well water), emissions to the environment, and the output(s) produced.
Material contribution Contribution of any one source of GHG emissions to a product amounting to more than one percent of its total anticipated life cycle GHG emissions. A materiality threshold of one percent has been established to ensure that minor sources of life cycle GHG emissions do not require the same treatment as more significant sources.
Mohair Fine, hairy fibre produced by an Angora goat.
Multi-functionality If a process or facility provides more than one function (i.e. it delivers several goods and/or services) or “co-products”, it is “multi-functional”. In these situations, all inputs and emissions linked to the process must be partitioned between the product of interest and the other co-products in a principled manner.
Normalization After the characterization step, normalization is an optional step in which the impact assessment results are multiplied by normalization factors that represent the overall inventory of a reference unit (e.g. a whole country or an average citizen). Normalized impact assessment results express the relative shares of the impacts of the analysed system in terms of the total contributions to each impact category per reference unit. When displaying the normalized impact assessment results of the different impact topics next to each other, it becomes evident which impact categories are affected most and least by the analysed system. Normalized impact assessment results reflect only the contribution of the analysed system to the total impact potential, not the severity/relevance of the respective total impact. Normalized results are dimensionless, but not additive.
Offal The internal organs of the body removed from the butchered animal (not included in a carcass).
Offsetting Mechanism for claiming a reduction in GHG emissions associated with a given process or product by removing or preventing the release of GHG emissions in an unrelated process or product.
Output Product, material or energy that leaves a unit process.
Ozone depletion Impact category that accounts for the degradation of stratospheric ozone due to emissions of ozone-depleting substances, for example, long-lived chlorine and bromine-containing gases (e.g. CFCs, HCFCs, Halons).
Packing Process of packing products in the production or distribution stages.
Primary data Data directly measured or collected data representative of processes at a specific facility or for specific processes within the product supply chain.
Primary packaging materials
Items other than containers/packaging, which reach consumers (including additives for preservation).
Process centre A facility where products are repackaged into smaller units without additional processing in preparation for retail sale.
xvi
Product parts Cuts of meat for retail sale (e.g., breast/thigh meat, wings, livers).
Product(s) Any good(s) or service(s). Services have tangible and intangible elements.
Proxy data Secondary LCI data or unit process used to represent a similar unit process in the product system. For example, using a Chinese unit process for electricity production in an LCA for a product produced in Viet Nam.
Ram An uncastrated (entire) male sheep.
Raw material Primary or secondary material used to produce a product.
Reference flow Quantity of a material from a unit process in the supply chain that is required to produce the functional unit.
Removal The sequestration or absorption of GHG emissions from the atmosphere, which most typically occurs when CO2 is absorbed by biogenic materials during photosynthesis.
Rendering A process that converts waste animal tissue into stable, value-added materials.
Replacement rate The percentage of adult animals in the herd replaced by younger adult animals.
Residual Material leaving the system in the condition as it created in the process, but which has a subsequent use. There may be value-added steps beyond the system boundary, but these activities do not impact the system calculations. Materials with economic value are not considered residual.
Ruminant Any of various even-toed, hoofed mammals of the suborder Ruminantia. Ruminants usually have a stomach divided into four compartments (one of which is called a rumen), and chew a cud consisting of regurgitated, partially digested food. Ruminants include cattle, sheep, goats, deer, giraffes, antelopes and camels.
Scouring Treating textiles in aqueous or other media to remove natural fats, waxes, proteins and other constituents, as well as dirt, oil and other impurities.
Secondary data Information obtained from sources other than direct measurement of the inputs/outputs (or purchases and emissions) from processes included in the life cycle of the product (PAS 2050:2011, 3.41). Secondary data are used when primary data are not available or it is impractical to obtain primary data. Some emissions, such as methane from litter management, are calculated from a model, and are therefore considered secondary data.
Secondary packaging materials
Containers/packaging and materials, which are used in raw materials acquisition, production and distribution but which do not reach consumers
Sink A natural or artificial reservoir that accumulates and stores some carbon-containing chemical compound for an indefinite period.
System boundary Set of criteria specifying which unit processes are part of a product system (life cycle).
System expansion Expanding the product system to include additional functions related to co-products.
Tallow Rendered fat from sheep.
Tier-1 method Simplest method that relies on single default emission factors (e.g. kg methane per animal).
xvii
1
Tier-2 method A more complex approach that uses detailed country-specific data (e.g. gross energy intake and methane conversion factors for specific livestock categories).
Tier-3 method Method based on sophisticated mechanistic models that account for multiple factors such as diet composition, product concentration from rumen fermentation, and seasonal variation in animal and feed parameters.
Uncertainty analysis Procedure to assess the uncertainty introduced into the results of a PEF study due to data variability and choice-related uncertainty.
Unit process The smallest unit of analysis in a life cycle assessment. It is used to keep an account of the exchanges necessary to produce the reference flow of the product output from the operation being modeled. In preparing the LCI model of the system, many unit processes are linked, with the output from each one potentially used as an input exchange for another. Thus the LCI model is a representation of the supply chain necessary to produce the product from the top-most unit process in the model.
Upstream emissions Emissions associated with processes that occur in the life cycle of a product prior to the processes owned, operated or controlled by the organization undertaking the assessment.
Volatile Solids (VS) Volatile solids (VS) are the organic material in livestock manure and consist of both biodegradable and non-biodegradable fractions. The VS content of manure equals the fraction of the diet consumed that is not digested and thus excreted as faecal material which, when combined with urinary excretions, constitutes manure.
Waste Materials for disposal that have no value, and may require further processing as part of the management of the materials. See also: Residual.
Water consumption Water withdrawal minus the return flow to rivers, lakes, aquifers and sea.
Weaning Removal of lambs or kids from their mothers, usually at about 10–16 weeks.
Wool The outer coat of sheep consisting of short curly hairs.
1
PART 1: 1
OVERVIEW AND GENERAL PRINCIPLES 2
2
1 INTENDED USERS AND OBJECTIVES 1
The methodology and guidance developed here can be used by stakeholders in all countries and across 2
the entire range of small ruminant production systems. In developing the guidelines, it was assumed 3
that the primary users will be individuals or organizations with a good working knowledge of life 4
cycle assessment. The main purpose of the guidelines is to provide sufficient definition of calculation 5
methods and data requirements to enable consistent application of LCA across differing small 6
ruminant supply chains. 7
This guidance is relevant to a wide array of livestock stakeholders including: 8
Livestock producers who wish to develop inventories of their on-farm resources and assess the 9
performance of their production systems. 10
Supply chain partners such as feed producers, farmers and processors seeking a better 11
understanding of the environmental performance of products in their production processes. 12
Policy makers interested in developing accounting and reporting specifications for livestock 13
supply chains. 14
The benefits of this approach include: 15
Use of recognized, robust and transparent methodology developed to take account of the 16
nature of small ruminant supply chains; 17
Identification of supply chain hotspots and opportunities to improve and reduce environmental 18
impact; 19
Identification of opportunities to increase efficiency and productivity; 20
Ability to benchmark performance internally or against industry standards; 21
Supporting reporting and communication requirements; and 22
Raising awareness and supporting action on environmental sustainability. 23
24
2 SCOPE 25
2.1 Environmental impact categories addressed in the guidelines 26
These guidelines cover only the following environmental impact categories: climate change, and fossil 27
energy demand. This document does not provide support towards the assessment of comprehensive 28
environmental performance, nor to the social or economic aspects or small ruminant supply chains. 29
The environmental impact categories were selected by the Technical Advisory Group (TAG) 30
members, based on the following criteria: 31
3
Relevance for the feed and livestock sectors as well as to the agendas of governments, 1
intergovernmental organizations, non-government organizations, civil society and the private 2
sector; 3
Agreement in the LCA community on the validity of the impact categorization model 4
(scientific consensus); 5
Quality and availability of characterization factors; 6
Local versus global level of impact. 7
The LEAP Animal Feed Guidelines cover additional impact categories: acidification, eutrophication 8
and land use. These categories may be reported for the life cycle stages of small ruminant products. It 9
is intended that in future these guidelines will be updated to include multiple categories. 10
Biodiversity loss, phosphorus depletion, water consumption, depletion of marine resources, soil 11
degradation, and eco-toxicity are other environmental impacts that the TAG considered highly 12
relevant but for which no consensual quantification techniques are available. For this reason, they 13
could not be included in the guidelines. Human toxicity, ozone depletion, ionising radiation and 14
photochemical ozone formation were estimated to be less important impact categories. 15
In the LEAP Animal Feed Guidelines, GHG emission from direct land-use-change is analysed and 16
recorded separately from GHG emissions due to other sources. There are two reasons for doing this. 17
The first is a question of time frame because emissions attributed to land-use-change may have 18
occurred in the past or may be set to occur in the future. Secondly, there is much uncertainty and 19
debate about the best method for calculating direct land-use-change. 20
Regarding land use, the LEAP Animal Feed Guidelines divided land areas into two categories: arable 21
land and grassland. Appropriate indicators were included in the guidelines as they provide important 22
information about the use of a finite resource (land) but also in view of the follow-on impacts on soil 23
degradation, biodiversity, carbon sequestration or loss, water depletion, etc. Nevertheless, users 24
wishing to specifically relate land use to follow-on impacts will need to collect and analyse additional 25
information on production practices and local conditions. 26
27
2.2 Application 28
Some flexibility in methodology is desirable to accommodate the range of possible goals and special 29
conditions arising in different sectors. This document strikes a pragmatic balance between flexibility 30
and rigorous consistency across scale, geographic location, and project goals. 31
A more strict prescription on the methodology, including allocation and acceptable data sources, is 32
required for product labelling or comparative performance claims. Users are referred to ISO 14025 for 33
more information and guidance on comparative claims of environmental performance. 34
4
These guidelines are generally based on the attributional approach to life cycle accounting. The 1
approach refers to process-based modelling, intended to provide a static representation of average 2
conditions. 3
Due to the limited number of environmental impact categories covered here, results should be 4
presented in conjunction with other environmental metrics to understand the wider environmental 5
implications, either positive or negative. It should be noted that comparisons between final products 6
should only be based on full life cycle assessment. Users of these guidelines shall not employ results 7
to claim overall environmental superiority of some small ruminant production systems and products. 8
The methodology and guidance developed in the LEAP Partnership is not intended to create barriers to 9
trade or contradict any WTO requirements. 10
These guidelines have been developed with a focus on sheep and goat production. Their application to 11
other small ruminant species is possible, however, there may be specific circumstances for other 12
species not covered in this document; for example, the co-production of velvet (antlers) and meat by 13
deer would require additional consideration regarding questions of allocation methodology. 14
15
3 STRUCTURE AND CONVENTIONS 16
3.1 Structure 17
This document adopts the main structure of ISO 14040:2006 and the four main phases of Life Cycle 18
Assessment – goal and scope definition, inventory analysis, impact assessment, and interpretation. 19
Figure 1presents the general relationship between the phases of an LCA study defined by ISO 20
14040:2006 and the steps needed to complete a GHG inventory in conformance with this guidance. 21
Part 2 of this methodology sets out the following: 22
Section 7 outlines the operational areas to which these guidelines apply. 23
Section 8 includes requirements and guidance to help users define the goals and scope, and 24
system boundary of an LCA. 25
Section 9 presents the principles for handling multiple co-products and includes requirements 26
and guidance to help users select the most appropriate allocation method to address common 27
processes in their product inventory. 28
Section 10 presents requirements and guidance on the collection and assessment of the quality 29
of inventory data as well as on identification, assessment, and reporting on inventory 30
uncertainty. 31
Section 11 outlines key requirements, steps, and procedures involved in quantifying GHG and 32
other environmental impact inventory results in the studied supply chain. 33
5
Section 1212 provides guidance on interpretation and reporting of results and summarizes the 1
various requirements and best practice in reporting. 2
A glossary intended to provide a common vocabulary for practitioners has been included. Additional 3
information is presented in the appendices. 4
5
FIGURE 1: MAIN LIFE CYCLE STEPS IN THE SMALL RUMINANT SUPPLY CHAIN 6
7
Users of this methodology should also refer to other relevant guidelines where necessary and 8
indicated. The LEAP small ruminants guidelines are not intended to stand alone but are meant to be 9
used in conjunction with the LEAP Animal Feed Guidelines. Relevant guidance developed under the 10
LEAP Partnership but contained in other documents will be specifically cross-referenced to enable 11
ease of use. For example, specific guidance for calculating associated emissions for feed is contained 12
in the LEAP Animal Feed Guidelines . 13
6
3.2 Presentational conventions 1
These guidelines are explicit in indicating which requirements, recommendations, or permissible or 2
allowable options that users may choose to follow. 3
The term “shall” is used to indicate what is required for an assessment to conform to these guidelines. 4
The term “should” is used to indicate a recommendation, but not a requirement. 5
The term “may” is used to indicate an option that is permissible or allowable. 6
Commentary, explanations and general informative material (e.g. notes) are presented in footnotes, 7
and do not constitute a normative element. 8
Examples illustrating specific areas of the guidelines are presented in boxes. 9
10
4 ESSENTIAL BACKGROUND INFORMATION AND PRINCIPLES 11
4.1 A brief introduction to LCA 12
Life cycle assessment (LCA) is recognized as one of the most important methods developed to assess 13
the environmental impact of products and processes. LCA can be used as a decision support tool 14
within environmental management. ISO 14040:2006 defines LCA as a “compilation and evaluation of 15
the inputs, outputs and the potential environmental impacts of a product system throughout its life 16
cycle”. In other words, LCA provides quantitative, confirmable, and manageable process models to 17
evaluate production processes, analyse options for innovation, and improve understanding of complex 18
systems. LCA can identify processes and areas where process changes stemming from research and 19
development can significantly contribute to reduce environmental impacts. According to 20
ISO14040:2006, LCA consist of four phases: 21
Goal and scope definition – including appropriate metrics (e.g. greenhouse gas emissions, 22
water consumption, hazardous materials generated, and/or quantity of waste); 23
Life cycle inventories (collection of data that identify the system inputs and outputs and 24
discharges to the environment); 25
Performance of impact assessment (application of characterization factors to the LCI 26
emissions which normalizes groups of emissions to a common metric such as global warming 27
potential reported in CO2 equivalents); 28
Analysis and interpretation of results. 29
7
4.2 Environmental impact categories 1
Life Cycle Impact Assessment (LCIA) aims at understanding and evaluating the magnitude and 2
significance of potential environmental impacts for a product system throughout the life cycle of the 3
product (ISO 14040:2006). The selection of environmental impacts is a mandatory step of LCIA and 4
this selection shall be justified and consistent with the goal and scope of the study (ISO 14040:2006). 5
Impacts can be modelled at different levels in the environmental cause-effect chain linking elementary 6
flows of the life cycle inventory to midpoint and areas of protection (Figure 2). 7
8
FIGURE 2: ENVIRONMENTAL CAUSE‐EFFECT CHAIN AND CATEGORIES OF IMPACT 9
10
Source: adapted from ILCD, 2010. 11 12
A distinction must be made between midpoint impacts (which characterize impacts somewhere in the 13
middle of the environmental cause-effect chain), and endpoint impacts (which characterize impacts at 14
the end of the environmental cause-effect chain). Endpoint methods provide indicators at, or close to, 15
an area of protection. Usually three areas of protection are recognized: human health, ecosystems, and 16
natural resources. The aggregation at endpoint level and at the areas of protection level is an optional 17
phase of the assessment according to ISO 14044:2006. 18
8
Climate change is an example of a midpoint impact category. The results of the Life Cycle Inventory 1
are the amounts of greenhouse gas emissions per functional unit. Using a characterization model and a 2
characterization factor such as the Global Warming Potential for each gas these results can be 3
expressed under the same midpoint impact category indicator, i.e. kilograms of CO2 equivalents per 4
functional unit. 5
These guidelines provide guidance on a selection of midpoint impact categories and indicators (Figure 6
2). They do not, however, provide guidance or recommendations regarding endpoint methods. 7
The LEAP Animal Feed Guidelines include some more categories and related methodologies 8
(Figure 2). These guidelines do not describe methodologies for other resource use and environmental 9
impact categories, but some relevant methodologies are described for the following: 10
land use or land occupation (which should be further subdivided into land suitable or 11
unsuitable for arable production, since it is important to recognize the potential of small 12
ruminants for utilizing land that is otherwise incapable of growing arable crops for direct 13
human consumption); 14
water consumption, accounting for blue water use and water scarcity (e.g. Ridoutt and Pfister, 15
2010; for water footprint methodologies see ISO/TC 14046, 2013); 16
resource depletion of non-renewable resources such as minerals and fossil fuels (e.g. Guinée 17
et al., 2002). 18
eutrophication (e.g. the eutrophication potential method of Guinée et al., 2002, or separate 19
eutrophication terrestrial and aquatic methodologies, as in Goedkoop et al., 2009). 20
21
4.3 Normative references 22
The following referenced documents are indispensable in the application of this methodology and 23
guidance. 24
ISO 14040:2006 Environmental management – Life cycle assessment – Principles and 25
framework 26
These standards give guidelines on the principles and conduct of LCA studies providing 27
organizations with information on how to reduce the overall environmental impact of their 28
products and services. ISO 14040:2006 define the generic steps which are usually taken when 29
conducting an LCA and this document follows the first three of the four main phases in 30
developing an LCA (Goal and scope, Inventory analysis, Impact assessment and 31
Interpretation). 32
9
ISO14044:2006 Environmental management – Life cycle assessment – Requirements and 1
guidelines 2
ISO 14044:2006 specifies requirements and provides guidelines for life cycle assessment 3
including: definition of the goal and scope of the LCA, the life cycle inventory analysis (LCI) 4
phase, the life cycle impact assessment (LCIA) phase, the life cycle interpretation phase, 5
reporting and critical review of the LCA, limitations of the LCA, relationship between the 6
LCA phases, and conditions for use of value choices and optional elements. 7
ISO 14025:2006 Environmental labels and declarations – Type III environmental declarations 8
– Principles and procedures 9
ISO 14025:2006 establishes the principles and specifies the procedures for developing Type 10
III environmental declaration programmes and Type III environmental declarations. It 11
specifically establishes the use of the ISO 14040 series of standards in the development of 12
Type III environmental declaration programmes and Type III environmental declarations. 13
Type III environmental declarations are primarily intended for use in business-to-business 14
communication, but their use in business-to-consumer communication is not precluded under 15
certain conditions. 16
ISO/TS 14067:2013 Greenhouse gases – Carbon footprint of products – Requirements and 17
guidelines for quantification and communication 18
ISO/TS 14067:2013 specifies principles, requirements and guidelines for the quantification 19
and communication of the carbon footprint of a product (CFP), based on ISO 14040 and ISO 20
14044 for quantification and on environmental labels and declarations (ISO 14020, ISO 14024 21
and ISO 14025) for communication. 22
WRI/WBCSD (2011) Product Life Cycle Accounting and Reporting Standard 23
The GHG Protocol from the World Resources Institute & World Business Council for 24
Sustainable Development (WRI/WBCSD) provides a framework to assist users in estimating 25
the total GHG emissions associated with the life cycle of a product. It is broadly similar in its 26
approach to the ISO standards, although it lays more emphasis on analysis, tracking changes 27
over time, reduction options and reporting. Like PAS2050, this standard excludes impacts 28
from production of infrastructure, but whereas PAS2050 includes ‘operation of premises’ such 29
as retail lighting or office heating, the GHG Protocol does not. 30
British Standards Institution PAS 2050:2011 Specification for the assessment of life cycle 31
greenhouse gas emissions of goods and services 32
PAS 2050:2011(BSI 2011) is a Publicly Available (i.e. not standard) Specification. A UK 33
initiative sponsored by the Carbon Trust and Defra, PAS 2050 was published through the 34
British Standards Institution (BSI) and uses BSI methods for agreeing a Publicly Available 35
Specification. It is targeted at applying LCA over a wide range of products in a consistent 36
manner for industry users, focusing solely on the carbon footprint indicator. PAS 2050 has 37
10
many elements in common with the ISO 14000 series methods but also a number of 1
differences, some of which limit choices for analysts (e.g. exclusion of capital goods and 2
setting materiality thresholds). 3
4
4.4 Guiding principles 5
Five guiding principles support users in their application of this sector-specific methodology. These 6
principles are consistent across the methodologies developed within the LEAP Partnership. They 7
apply to all the steps, from goal and scope definition, data collection and LCI modelling through to 8
reporting. Adhering to these principles ensures that any assessment made in accordance with the 9
methodology prescribed is carried out in a robust and transparent manner. The principles can also 10
guide users when making choices not specified by the guidelines. 11
The principles are adapted from the WBCSD-WRI’s Greenhouse Gas Protocol Product Life Cycle 12
Accounting and Reporting Standard (2011), the BSI PAS 2050:2011, the ILCD Handbook and ISO/TS 13
14067:2013 and are intended to guide the accounting and reporting of environment impacts categories. 14
Accounting and reporting of GHG emissions and other environmental impacts from small ruminant 15
supply chains shall accordingly be based on the following principles: 16
Relevance 17
Data, accounting methodologies and reporting shall be appropriate to the decision-making needs of the 18
intended users. Information should be reported in a way that is easily understandable to the intended 19
users. 20
Completeness 21
All product life cycle GHG emissions, removals and sinks, and other environmental criteria within the 22
specified – system and temporal – boundaries under study, shall be reported. Any specific exclusion 23
shall be disclosed and justified. 24
Consistency 25
Consistent methodologies, data and assumptions shall be used throughout the assessment to allow for 26
meaningful comparisons and reproducibility of the outcomes over time. Any changes to the data, 27
boundaries, assumptions, methods, or any other relevant factors shall be reported and documented.28
11
Accuracy 1
Bias and uncertainties shall be reduced as far as practicable. Sufficient accuracy shall be achieved to 2
enable intended users to make decisions with reasonable confidence as to the reliability and integrity 3
of the reported information. 4
Transparency 5
In external communications, sufficient information shall be disclosed and appropriate references made 6
to allow third parties to verify all data, calculations and assumptions, and intended users to make 7
associated decisions with confidence. A clear audit trail shall address all the relevant issues in a factual 8
and coherent manner. 9
10
5 LEAP AND THE PREPARATION PROCESS 11
LEAP is a multi-stakeholder initiative launched in July 2012 with the goal of improving the 12
environmental performance of livestock supply chains. Hosted by the Food and Agriculture 13
Organization of the United Nations, LEAP brings together the private sector, governments, civil 14
society representatives and leading experts who have a direct interest in the development of science-15
based, transparent and pragmatic guidance to measure and improve the environmental performance of 16
livestock products. 17
Demand for livestock products is projected to grow 1.3% per annum until 2050, driven by global 18
population growth and increasing wealth and urbanization (Alexandratos and Bruinsma, 2010). 19
Against the background of climate change and increasing competition for natural resources, this 20
projected growth places significant pressure on the livestock sector to perform in a more sustainable 21
way. The identification and promotion of the contributions that the sector can make towards more 22
efficient use of resource and better environmental outcomes is also important. 23
Currently, many different methods are used to assess the environmental impacts and performance of 24
livestock products. This causes confusion and makes it difficult to compare results and set priorities 25
for continuing improvement. With increasing demands in the marketplace for more sustainable 26
products there is also the risk that debates about how sustainability is measured will distract people 27
from the task of driving real improvement in environmental performance. And there is the danger that 28
labelling or private standards based on poorly developed metrics could lead to erroneous claims and 29
comparisons. 30
The LEAP Partnership addresses the urgent need for a coordinated approach to developing clear 31
guidelines for environmental performance assessment based on international best practices. The scope 32
of LEAP is not to propose new standards but to produce detailed guidelines that are specifically 33
relevant to the livestock sector, and refine guidance as to existing standards. LEAP is a multi-34
12
stakeholder partnership bringing together the private sector, governments and civil society. These 1
three groups have an equal say in deciding work plans and approving outputs from LEAP, thus 2
ensuring that the guidelines produced are relevant to all stakeholders, widely accepted and supported 3
by scientific evidence. 4
With this in mind, the first three technical advisory groups (TAGs) of LEAP were formed in early 5
2013 to develop guidelines for assessing the environmental performance of small ruminants (goats and 6
sheep), animal feeds and poultry supply chains. 7
The work of LEAP is challenging but vitally important to the livestock sector. The diversity and 8
complexity of livestock farming systems, products, stakeholders and environmental impacts can only 9
be matched by the willingness of the sector’s practitioners to work together to improve performance. 10
LEAP provides the essential backbone of robust measurement methods to enable assessment, 11
understanding and improvement in practice. More background information on the LEAP Partnership 12
can be found at www.fao.org/partnerships/leap/en/ 13
14
5.1 Development of sectorspecific guidelines 15
Sector-specific guidelines to assessing the environmental performance of the livestock sector are a key 16
aspect of the LEAP Partnership work programme. Such guidelines take into account the nature of the 17
livestock supply chain under investigation and are developed by a team of experts with extensive 18
experience in life-cycle assessment and livestock supply chains. 19
The benefit of a sector-specific approach is that it gives guidance on the application of life-cycle 20
assessment to users and provides a common basis from which to evaluate resource use and 21
environmental impacts. 22
Sector-specific guidelines may also be referred to as supplementary requirements, product rules, sector 23
guidance, product category rules or product environmental footprint category rules – although each 24
programme will prescribe specific rules to ensure conformity and avoid conflict with any existing 25
parent standard. 26
The first set of sector-specific guidelines addresses small ruminants, poultry and animal feeds. The 27
former two place emphasis on climate-related impacts, while the LEAP Animal Feed Guidelines 28
address a broader range of environmental categories. LEAP is also considering developing guidance 29
for the assessment of other animal commodities and wider environmental impacts such as biodiversity, 30
water and nutrients. 31
32
13
5.2 The Small ruminants TAG and the preparation process 1
The Small ruminants TAG of the LEAP Partnership was formed at the start of 2013. The team 2
included selected nine experts in small ruminant supply chains as well as leading LCA researchers and 3
experienced industry practitioners. Their backgrounds, complementary between products, systems and 4
regions, allowed them to understand and address different interest groups and so ensure credible 5
representation. The TAG was led by Dr Stewart Ledgard of AgResearch, New Zealand. 6
The role of the TAG was to: 7
Review existing methodologies and guidelines for assessment of GHG emissions from 8
livestock supply chains and identify gaps and priorities for further work; 9
Develop methodologies and sector specific guidelines for the life cycle assessment of GHG 10
emissions from small ruminant supply chains; and 11
Provide guidance on future work needed to improve the guidelines and encourage greater 12
uptake of life-cycle assessment of GHG emissions from small ruminant supply chains. 13
The TAG met for its first workshop on 12–14 February 2013. The TAG continued to work via emails 14
and teleconferences before meeting for a second workshop, which took place on 5–7 September 2013 15
in Rome. The eight experts were drawn from six countries: Australia, France, Germany, Malaysia, 16
New Zealand and the United Kingdom. 17
As a first step, existing studies and associated methods were reviewed by the TAG to assess whether 18
they offered a suitable framework and orientation for a sector-specific approach. This avoids 19
confusion and unnecessary duplication of work through the development of potentially competing 20
standards or approaches. The review also followed established procedures set by the overarching 21
international guidance sources listed in section 4.3. 22
Our selection of documents for background review in support of the development of these guidelines 23
was driven by the availability of full LCA studies in the small ruminant sector. Our interest was in 24
determining the range of methodological choices that have been used. Our intention was to evaluate as 25
broadly as possible and therefore included peer-reviewed articles, conference proceedings and 26
technical reports. These sources allowed an evaluation of the methodological consistencies and 27
differences for global systems. 28
The TAG identified 12 studies addressing aspects of the small ruminant supply chain, with eight 29
covering only the cradle-to-farm-gate, four covering the cradle-to-retail gate, and one covering a 30
whole life cycle (meat only). All 12 studies focused on sheep, with one also covering goats. 31
Appendix 1 presents a review of these studies. As a result, it was concluded that no existing approach 32
or study set out fully comprehensive guidance for quantifying GHG emissions and energy use across 33
the supply chain, and that the TAG would need to undertake further work to reach consensus on more 34
detailed guidance. This task built on initial work on a methodology for carbon footprinting of lamb 35
14
(cradle-to-farm-gate) by LCA researchers (including some in this TAG), which was supported by the 1
International Meat Secretariat and Beef + Lamb New Zealand. 2
3
5.3 Period of validity 4
It is intended that these guidelines will be periodically reviewed to ensure the validity of the information 5
and methodologies on which they rely. At the time of development, no mechanism is in place to ensure 6
such review. The user is invited to visit the LEAP website (www.fao.org/partnerships/leap) to obtain the 7
latest version. 8
9
6 BACKGROUND INFORMATION ON SMALL RUMINANT SUPPLY CHAINS 10
6.1 Background and context 11
The world populations of goats and sheep in 2011 were 876 and 1 043 million, respectively. A 12
breakdown of their distribution across the main countries shows a dominance of goats in Africa and 13
Asia, although the dominance of specific products varies with most meat production in China, while 14
most milk production occurs in India (Table 6 in Appendix 2). Similarly, for sheep, China is the 15
largest producer of meat, wool and milk, although Australia and New Zealand are the next largest 16
producers of meat and wool (Table 7 in Appendix 2). The various world regions also show differences 17
in terms of their temporal trends in sheep and goat production. The production of both milk and meat 18
from sheep and goats has increased significantly during the past 20 years in Asia, while the increase 19
was lower in Africa (Figure 3). However, production trends were stable to declining in Europe and 20
Oceania. 21
15
FIGURE 3: TRENDS IN THE GLOBAL PRODUCTION OF SHEEP AND GOATS (A) MEAT AND (B) MILK ACROSS WORLD 1 REGIONS 2
3
Note: Production is summed between the two species (FAOSTAT, 2013). Regions with low production: meat production for 4 North and South America went from 0.8 and 0.4 million t in 1980 to 0.4 and 0.09 million t in 2011, respectively; milk production 5 for South America and Oceania went from 182 and 0.03 x103 t in 1980 to 254 and 0.04 x103 t in 2011, respectively. 6 7
Both species present a bewildering mix of breeds and play valuable multi-functional roles, especially in 8
small farm systems. Their preferred environments are somewhat different, with goats being more 9
heavily concentrated in semi-arid and arid areas, while sheep thrive best in cooler environments. Goats 10
and sheep play an important socio-economic role in many rural areas. They are capable of utilizing low-11
quality fibrous feeds (goats more so than sheep) and are highly valued for the multiple products they 12
produce. These include edible products such as meat and milk, and non-edible products such as manure, 13
hides and skins, and natural fibre (mohair, cashmere or wool). With the exception of larger farms where 14
one or the other species may be reared intensively for a particular product, in most small farms and low-15
input systems, both species may be reared together (often for purposes of diversification). 16
17
6.2 Diversity of small ruminant production systems 18
The agro-ecosystem conditions (climatic, edaphic, biotic) determine which plants exist or can 19
potentially be grown. This in turn determines the quantity, quality and distribution of the feed base, 20
which governs the development of potential animal production systems. Small ruminant production 21
occurs worldwide across a range of Agro-Ecological Zones (AEZ) and presents a wide diversity of 22
systems with different intensities and objectives of production (Figure 4). While a range of constraints 23
to production of small ruminants exist across the AEZs (Devendra et al., 2005), sheep and, in 24
particular, goats are also well adapted to a wide range of conditions and AEZs often unsuitable for 25
alternative food production systems. Additionally, small ruminant production in these landscapes can 26
result in positive environmental benefits through nutrient cycling and maintenance of biodiversity. 27
16
FIGURE 4: GLOBAL DISTRIBUTION OF (A) SHEEP AND (B) GOATS FROM THE TWO MAIN PRODUCTION SYSTEMS – 1 GRASSLAND BASED AND MIXED 2
3
Note: The color (see legend) indicates the dominant production system and the number of animals in each grid cell. Source: 4 Gerber et al, 2013. 5
17
Examples of the diversity of sheep or goat production systems are as follows. 1
1. Intensive production systems for meat or milk as the main product with animals housed 2
permanently or through most of the year. Feed supply can be brought in totally from arable 3
crops or from cut-and-carry pasture and cultivated improved forages. The system is common 4
in humid regions where feed is generally more plentiful. 5
2. Extensive to intensive systems with animals reared predominantly on pastures in confined 6
farms. These may include animals housed for part of the year through to some animals 7
(e.g. lambs) being housed throughout their life, with concentrates being fed during the 8
confinement period. Main products may include meat, milk or fibre. 9
3. Extensive systems with animals managed communally for grazing and fed on native forages 10
and residues from crops or trees. Main products may include meat, milk or fibre (manure may 11
also be a useful co-product). 12
4. Very extensive systems where animals are grazed on large areas of unproductive and 13
marginal lands, including rangelands, forest areas and roadsides. Very low annual rainfall 14
produces a sparse feed-resource base where animals have to seek feed. Inadequate control of 15
numbers can lead to overgrazing and damage to the environment. This system is very common 16
in semi-arid and arid agro-ecological zones. 17
5. Nomadic and transhumance systems that involve regular movements of whole flocks (along 18
with families). These systems are found in agro-ecosystems where crop production is not 19
possible. Grazing and water availability are the main drivers of these movements, which can 20
involve very large flocks. 21
6. Rural landless production system. Several million poor farmers are found in landless small 22
ruminant production systems, especially within systems 3 to 5 above. The poorest are found in 23
vulnerable semi-arid and arid AEZs. In this system, small ruminants play a vital role in 24
ensuring survival through meat and milk, and some income. 25
7. Sylvopastoral systems where small ruminant production is integrated with tree cropping. 26
Residues from trees are often used as feed. These integrated systems are a good example of 27
diversification, which is mainly driven by seasonality and risk. 28
29
6.3 Diversity of small ruminant value chains 30
Value chains play an important role in linking the various stages from production to consumption and 31
waste, including the many services involved. In this context, a life cycle assessment (LCA) approach 32
is appropriate to account for the many stages of resource use and environmental emissions throughout 33
the value chain from raw materials to production, transportation, processing, consumption and waste 34
18
management. A value chain approach enables identification of potential factors for improvement 1
throughout the life cycle. 2
In a number of developing countries, small farm owners encounter major problems in coping with a 3
range of difficulties in the face of the complexity and general inefficiency of prevailing marketing 4
chains. Foremost among these is access to a marketing chain. At present in developing countries, 5
inadequate access to market outlets and weak marketing arrangements represent a major constraint to 6
the production to consumption systems and to smallholder owners and producers of sheep and goats. 7
The market chain involves rural, peri-urban, urban and international markets, and a major challenge 8
lies in ways to link small farmers with these markets and marketing systems. In Asia, village slaughter 9
centres can be important for increased access to marketing chains by farmers and include socio-10
economic elements. Rural markets are especially important to rural communities and their households, 11
and are also used for the sale of live animals for slaughter in urban areas. Without improved marketing 12
and transport systems, the prevailing systems constitute major constraints to the sale of animals and 13
products from small farms. 14
In contrast, in intensive production systems in developed countries, linked value chains are prevalent. 15
They are generally associated with large processing facilities and attempt to gain greater value from 16
the many co-products from small ruminants. 17
The processing of products from small ruminants can potentially involve many complex stages with 18
multiple end products. These guidelines extend only to primary processing, of which there are a 19
diversity of primary processing systems: 20
Specialist abattoirs disassemble animals into a very wide range of meat products and co-21
products. The latter include hides (for leather), tallow (e.g. for soap, biofuel), internal organs 22
and meat waste (for pet food), blood (e.g. for pharmaceutical products), fibre and renderable 23
material (e.g. for fertilizer). 24
Specialist milk processing plants produce a wide range of basic products including cheese, 25
yoghurt, whey and dried milk. 26
Specialist fibre scouring plants wash and clean the fibre and may produce co-products such as 27
lanolin. 28
Some animals are sold for “backyard” slaughter (sometimes called “wet markets”) primarily 29
for meat products. 30
Village slaughter centres (especially in Asia) are associated with the slaughter of a relatively 31
small number of animals to provide meat to villagers. At these centres, the offal and skins are 32
generally sold on and processed at other specialist sites. 33
19
Alternatively, primary processed products (e.g. packaged cuts of meat from abattoirs, wool for use in 1
insulation of houses) may be sent directly to wholesale or retailers for direct sale to customers. It is 2
acknowledged that the various stages after primary processing, through to consumption and waste, 3
may result in significant use of energy and refrigerants with associated GHG emissions. However, data 4
requirements for these stages are often difficult to obtain and are usually derived from secondary data 5
in published reports. It was considered impractical to attempt to include these various stages after 6
primary processing in the current guidelines. However, a number of other Specifications or PCRs 7
account for secondary processing and subsequent stages for textiles (e.g. BSI, 2013) and meat 8
(e.g. Boeri, 2013). 9
A very wide range of secondary processing systems exist, but no attempt was made to account for 10
them in these guidelines. Examples include transforming meat into specialist cuts or to final processed 11
products (e.g. cooked lasagne pre-packed for retailers); carding and spinning fibre into yarns for 12
clothing or carpet production; and addition of further ingredients to basic milk products to produce 13
specialist products such as infant milk products. 14
15
6.4 Multifunctionality of small ruminant supply chains 16
Small ruminant production systems generate a range of goods and services, thus providing a wide 17
range of contributions towards economic, social, food security, and agronomic and ecological 18
objectives. 19
For many poor and vulnerable people, small ruminants play an important role in the four dimensions 20
of food security (availability, access, stability and utilization) and are crucial to nutrition, providing 21
high-quality proteins and a wide diversity of micronutrients. Where people have no access to banks 22
and other financial services, small ruminants allow them to store and manage wealth, and are thus an 23
important buffer in times of crisis. 24
In mixed crop-livestock systems, small ruminants often contribute to productivity, as manure is used 25
to fertilize the soil and maintain organic matter, and herds are used to control weeds. 26
Small ruminants also play an important role in cultural and religious events. Among these is the 27
Tabaski, celebrated by the Muslims of West and Central Africa, which requires the ritual sacrifice of 28
animals. 29
Finally, small ruminants can contribute effectively to the management of landscapes and preserved 30
ecosystems. In some areas (e.g. mountainous regions), small ruminants contribute to the cultural 31
landscape, providing ecosystem services through encroachment control, conservation of biodiversity, 32
and the presence of traditional agricultural activities and infrastructure. 33
20
6.5 Overview of global emissions from small ruminants 1
Globally, sheep and goats are responsible for about 6.5 percent of the livestock sector’s emissions 2
(475 million tonnes CO2-equivalents) (Gerber et al., 2013; Opio et al., 2013). The global average 3
GHG emission intensity of milk is lower for goats than for sheep (5.2 vs. 8.4 kg CO2-eq/kg product, 4
respectively), mainly because goats have higher milk yields on average at the global level. The 5
corresponding GHG emission intensity of meat is very similar between the two species (23.3 vs. 6
23.4 kg CO2-eq/kg product for goats and sheep, respectively). For both milk and meat, emission 7
intensity tends to be lower in developed than developing regions. However, this should not be 8
interpreted as suggesting an overall environmental superiority of developed country production 9
systems, as noted in Section 2.2. Enteric fermentation and feed production largely dominated the 10
sources of GHG emissions along the supply chains in these studies, accounting for 55 percent and 11
35 percent of emissions from small ruminants, respectively. In regions where natural fibre production 12
(wool, cashmere, mohair) is economically important, a substantial share of emissions can be attributed 13
to these products (especially when the economic value is used to allocate emissions between edible 14
and non-edible products). 15
Gerber et al. (2013) also show that emission intensities vary greatly between production units, even 16
within similar production systems, leaving much room for improvement. If the bulk of the world’s 17
small ruminant producers adopted technologies and practices already used by the most efficient 18
producers in terms of emission intensity, it would result in significant reductions in emissions. A 19
major driver of GHG emission intensity is the efficiency of feed conversion into product, which is 20
determined by potential animal productivity, as well as feed availability and quality through the year. 21
Manure management also has an important effect on GHG emissions. Opportunities for reducing GHG 22
emission intensity include improved animal breeding, feeding, health and reproduction. Management 23
practices to improve production and quality of feed sources, including through efficient use of manure 24
for better nutrient capture and recycling, can also enhance animal productivity. However, the potential 25
for reducing GHG emission intensity are dependent on local climatic and feed conditions. Indeed, in 26
some ecosystems, small ruminants may be one of the only options landholders have to utilize low 27
quality forage for production of protein for human consumption. In some grassland and rangeland 28
systems there is also potential for increased carbon sequestration in vegetation and soils. 29
21
PART 2: 1
METHODOLOGY FOR QUANTIFICATION OF GREENHOUSE 2
GAS EMISSIONS AND FOSSIL ENERGY DEMAND OF SMALL 3
RUMINANT PRODUCTS 4
22
7 DEFINITION OF PRODUCTS AND PRODUCTION SYSTEMS 1
This document is intended to provide guidelines for users to calculate the GHG emissions and 2
removals for small ruminant (sheep and goats) products over the key stages of the cradle-to-farm-gate 3
or the cradle-to-primary processing gate. The guidelines are based on use of an attributional life cycle 4
assessment (LCA) approach. It is expected that the primary users will be individuals or organizations 5
that have a good working knowledge of LCA. 6
7
7.1 Products description 8
These guidelines cover the cradle-to-primary processing gate and the main products generated 9
comprise one or more of: 10
meat products, with possible co-products such as tallow, skins and renderable material; 11
clean fibre (i.e. wool, mohair or cashmere) and possible minor lanolin co-product; 12
milk products such as cheese, yoghurt and milk powder, with possible co-products such as 13
whey. 14
These products are produced from a very diverse range of production systems around the globe. 15
16
7.2 Life cycle stages: modularity 17
A life cycle assessment of primary products can be conducted by dividing the production system into 18
modules that relate to the different life cycle stages. The three main stages can be summarized as feed 19
production (including feed processing, milling and storage), animal production (including animal 20
breeding) and primary animal processing (Section 8.4) (Figure 5). The feed production stage covers 21
the cradle-to-animal’s mouth stage and includes a range of feeds including processed concentrates, 22
forage crops, pastures, shrubs and trees (see the LEAP Animal Feed Guidelines). The animal 23
production stage covers the cradle-to-farm-gate stage and the main products generated are live 24
animals, fibre (e.g. wool for sheep and mohair or cashmere for goats) and/or milk. 25
26
FIGURE 5: MODULAR SCHEME OF SMALL RUMINANT PRODUCTION CHAINS 27
28
23
8 GOAL AND SCOPE DEFINITION 1
8.1 Goal of the LCA study 2
The first step required when initiating an LCA is to clearly set the goal or statement of purpose. This 3
statement describes the goal pursued and the intended use of results. Numerous reasons for performing 4
an LCA exist. LCAs can be used to serve the goal of GHG emission management by determining the 5
carbon footprint of products, understanding the GHG emission hotspots and prioritizing emissions-6
reduction opportunities along supply chains. LCAs provide detailed information on a product’s 7
environmental performance and can serve performance tracking goals as well as setting progress and 8
improvement targets. They could also be used to support reporting on the environmental impacts of 9
products, although these guidelines are not intended for comparison of products or labelling of 10
environmental performance. 11
It is therefore of paramount importance that the goal and scope be given careful consideration because 12
these decisions define the overall context of the study. A clearly articulated goal helps ensure that 13
aims, methods and results are aligned. For example, fully quantitative studies will be required for 14
benchmarking or reporting, but somewhat less rigour may be required for hotspot analysis. 15
Interpretation is an iterative process occurring at all steps of the LCA and ensuring that calculation 16
approaches and data match the goal of the study (Figure 1 and Section 12). Interpretation includes 17
completeness checks, sensitivity checks, consistency checks and uncertainty analyses. The conclusions 18
(reported or not) drawn from the results and their interpretation shall be strictly consistent with the 19
goal and scope of the study. 20
Seven aspects shall be addressed and documented during the goal definition (European Commission, 21
2010): 22
Subject of the analysis 23
Key properties of the assessed system: organization, location(s), dimensions, products, 24
sector, and position in the value chain; 25
Purpose of performing the study and decision-context; 26
Intended use of the results: will they be used internally for decision-making or shared 27
externally with third parties? 28
Limitations due to the method, assumptions, and choice of impact categories: in particular, 29
limitations to broad study conclusions associated with exclusion of impact categories shall 30
be addressed; 31
Target audience of the results; 32
Comparative studies to be disclosed to the public and need for critical review; 33
Commissioner of the study and other relevant stakeholders. 34
24
8.2 Scope of the LCA study 1
The scope is defined in the first phase of an LCA, as an iterative process with the goal definition. It 2
states the depth and breadth of the study. The scope shall identify the product system or process to be 3
studied, the functions of the system, the functional unit, the system boundaries, the allocation 4
principles, and the impact categories. The scope should be defined so that the breadth, depth and detail 5
of the study are compatible and sufficient to achieve the stated goal. While conducting an LCA of 6
livestock products, the scope of the study may need to be modified as information is collected so as to 7
reflect data availability and techniques or tools for filling data gaps. Specific guidance is provided in 8
the subsequent sections. It is also recognized that the scope definition will affect the data collection for 9
the life cycle inventory, as described in more detail in Section 10.1. 10
These guidelines refer only to two environmental impact category (climate change characterized 11
through GHG emissions, and reported as CO2 equivalent and fossil energy demand) and therefore 12
should not be used to provide an indicator of overall environmental effects of the production systems 13
and products. Care is therefore needed in the reporting and communication of the results of 14
assessments based on these guidelines to avoid misinterpretation of the scope and application of the 15
results. 16
17
8.3 Functional unit 18
The functional unit defines the form and units used to specify a product. Different functional units are 19
appropriate when different downstream system boundaries are specified (e.g. farm gate, processing 20
plant, retail, post-consumption to grave). Recommended functional units for different main product 21
types are given in Table 1. Where meat is the product type, the functional unit when the animal leaves 22
the farm shall be live-weight. At the stage of leaving the meat-processing plant (or abattoir) the 23
functional unit shall be the weight of product (meat-product-weight) destined for human 24
consumption. In many Western countries with commercial processing plants, the product weight has 25
traditionally been identified as carcass-weight at the stage of leaving the meat-processing plant. 26
Carcass-weight (sometimes called dead-weight) generally refers to the weight of the carcass after 27
removal of the skin, head, feet and internal organs, including the digestive tract (and sometimes some 28
surplus fat). However, these internal organs, for the most part, are edible. Red offals (e.g. liver, 29
kidney, heart) and green offals (e.g. stomach and intestines) are increasingly being harvested and 30
should be included. In developing countries, the meat-processing site may vary from processing plants 31
to “backyard” or cottage industry processing (sometimes termed “wet market”), and a higher 32
proportion of the animal may be harvested for human consumption. Note that the “product-weight” 33
includes bone retained within the animal parts for human consumption (primary processing plants for 34
small ruminants typically leave bone in many of the meat cuts). The relative bone content has been 35
25
estimated at approximately 18 percent of a sheep’s carcass-weight in UK studies (Kim Matthews, 1
EBLEX UK data). Ideally, the bone content of the total meat product would be defined, but this is 2
rarely measured. However, it shall be stated when the functional unit includes bone-in, and if the bone 3
content is outside the usual range for the carcass component it shall be described and an estimate of 4
the bone content provided. Where specific data for “product-weight” is not available, the cold carcass-5
weight shall be used and can be estimated from the live-weight using default values, based on a 6
summary of international data (Appendix 3). No distinction is made between different cuts of meat, 7
including edible offal, and they shall be treated as equivalent (with no specific allocation method used 8
for different cuts). An example of the relative content by weight of different meat cuts and co-products 9
is given in Box 4 in Section 11.6.3. 10
Where fibre is the main product type, the functional unit shall be greasy weight (as shorn off the 11
animals) at the farm gate or clean weight after it leaves a scouring plant. The scouring plant is the 12
only primary processing stage covered by these guidelines (see Section 8.4). 13
Where milk is the main product type, the functional unit shall be the weight of the milk as it leaves the 14
farm gate corrected for fat, protein and lactose content (energy-corrected milk). The latter 15
standardizes the milk after adjustment for differences associated with animal type, breed and 16
production. To provide comparison with dairy cow milk, the following equation from the IDF (2010a) 17
methodology is recommended for energy-corrected milk (ECM): 18
19
kg ECM = kg milk x (0.1226 x fat% + 0.0776 x true-protein% + 0.0621 x lactose%) 20
21
Where crude-protein % is used instead of true-protein %, the relevant multiplier is 0.0722 (instead of 22
0.0776). This equation standardizes the milk to 4 percent fat, 3.3 percent protein and 4.8 percent 23
lactose. Research indicates that lactose % can vary during the lactation season and with species 24
(e.g. Park and Haenlein, 2006), and therefore it is desirable to account for lactose % in the equation for 25
ECM. However, if data on lactose % are unavailable, a default value of 4.8 percent lactose shall be 26
used for sheep and goats. After the milk primary processing stage there are a wide range of possible 27
products and the appropriate functional unit reported shall be the weight of the specific product (milk 28
product weight). 29
26
TABLE 1: RECOMMENDED FUNCTIONAL UNITS FOR THE THREE DIFFERENT MAIN PRODUCT TYPES FROM 1 SMALL RUMINANTS ACCORDING TO WHETHER THEY ARE LEAVING THE FARM OR THE PRIMARY PRODUCT 2 PROCESSING GATE 3
4
8.4 System boundary 5
8.4.1 GENERAL/SCOPING ANALYSIS 6
The system boundary shall be defined following general supply chain logic, including all phases from 7
raw material extraction to the point at which the functional unit is produced. A full LCA would 8
include processing, distribution, consumption and final disposal; however, this guide does not cover 9
post primary processing stages in the supply chain. 10
The overall system boundary covered by these guidelines represents the cradle-to-primary processing 11
stages of the life cycle of the main products from small ruminants (Figure 6). It covers the main stages 12
of the cradle-to-farm-gate, transportation of animals to primary processor, and then to the primary 13
processing gate (e.g. to the output loading dock). 14
The modular approach outlined in Section 7.2 illustrates the three main stages of the cradle-to-primary 15
processing gate. The feeds stage is covered in detail in the associated LEAP Animal Feed Guidelines 16
document and covers the cradle-to-animal’s mouth stage for all feed sources (including raw materials, 17
inputs, production, harvesting, storage and feeding) and other feed-related inputs (e.g. milk powder for 18
feeding lambs/kids and nutrients directly fed to animals; covered in detail in Section 11.2). 19
The animal production stage covers all other inputs and emissions associated with animal production 20
and management not covered by the LEAP Animal Feed Guidelines. It is important to ensure that all 21
farm-related inputs and emissions are included in the feed and animal stages, and to avoid double 22
counting. The animal production stage includes accounting for breeding animals as well as those used 23
directly for meat/milk/fibre production. This may involve more than one farm if animals are traded 24
between farms prior to going to processing. 25
The primary processing stage shall be limited to the primary milk-processing factory, the scouring or 26
cleaning stage for fibre, and animal slaughter (backyard, village slaughter centre and abattoir) for meat 27
processing. It includes all transportation steps within the cradle-to-primary processing gate. 28
Main product type Cradle‐to‐farm‐gate Cradle‐to‐primary processing gate
Meat Live‐weight (kg) Meat product(s) (kg)
Fibre Greasy weight (kg) Clean weight (kg)
Milk Energy‐corrected milk (kg) Milk product(s) (kg)
27
The choice of basic milk, meat products and clean fibre as typical sector outputs is intended to provide 1
a point in the supply chain that has an analogue across the range of possible systems, geographies and 2
goals that may be encountered in practice. Basic milk and meat products may be used directly by the 3
consumer (particularly in developing countries) or may undergo further secondary processing with the 4
addition of other constituents to make more complex food products (e.g. sweetened fruit yoghurt, 5
lasagne). For fibre, a range of secondary processing options exist depending on the end product, for 6
example, yarn/fibre spinning, dyeing, knitting/weaving and garment-making; spinning and carpet or 7
rug making; or direct use as insulation or absorbent for contaminants such as oil spills. 8
Several available product category rules (PCRs) extend beyond the system boundary covered in these 9
guidelines and include the post-primary processing supply chain for meat (Boeri, 2013) and fibre 10
(Rossi, 2012). There are no PCRs for processed milk products from small ruminants, but there are for 11
dairy cow milk products (e.g. Sessa, 2013b). There are no PCRs for carpets, but some early LCA 12
publications exist for carpets (e.g. Petersen and Solberg, 2004; Potting and Blok, 1995) that illustrate 13
some of the non-fibre constituents and additional processes. There is a draft Specification under 14
development for textile products (BSI, 2013). 15
Figure 6 illustrates a range of co-products produced from the farm to primary processing gate, which 16
fall outside the system boundary covered by these guidelines. There are no PCRs related specifically 17
to these co-products. However, there are some relevant LCA publications for leather (Joseph and 18
Nithya, 2009; Mila i Canals et al., 1998; Mila i Canals et al., 2002), biofuel from tallow (Thamsiriroj 19
and Murphy, 2011), thermoplastic from blood meal (Bier et al., 2012) and products from rendering 20
animal processing byproducts (Ramirez et al., 2011). 21
22
a) Scoping analysis 23
Frequently a scoping analysis based on a relatively rapid assessment of the system can provide 24
valuable insight into areas which may require additional resources to establish accurate information 25
for the assessment. Scoping analysis can be conducted using secondary data to provide an overall 26
estimate of the system impact. Furthermore, based on existing literature reviews in the small ruminant 27
sector (Appendix 1), it is relatively clear that for production systems it is extremely important that the 28
following factors are assessed with high accuracy: the ration, the feed conversion efficiency, manure 29
production and management. Depending upon the particular operation under study, additional effects 30
may be observed. In the post-farm supply chain, energy efficiency at the processing and 31
manufacturing stages as well as an accurate assessment of transportation modes and distances is 32
important. 33
28
FIGURE 6: SYSTEM BOUNDARY DIAGRAM FOR THE LIFE CYCLE OF SHEEP OR GOATS COVERING THREE MAIN 1 PRODUCTS (FIBRE, MILK AND MEAT) 2
3
Note: The large box covering the cradle-to-primary processing gate represents the stages covered by guidelines in this document, 4 while the inner left box relating to land and feed is covered in the LEAP Animal Feed Guidelines. The encircled t symbol refers 5 to the main transportation stages. The terms in italics refer to functional units of products leaving several different stages. The 6 “sheep or goats” box may include up to several phases of movement of animals between different farms/areas/systems before 7 progress to primary processor. 8 9
8.4.2 CRITERIA FOR SYSTEM BOUNDARY 10
Material system boundary: Which entities and processes are included in the assessment? What is the 11
analysed company’s sphere of influence? Which entities and processes are excluded from the 12
assessment, and for what reasons? A flow diagram of all assessed processes should be drawn 13
indicating where processes were cut off. For the main transformation steps within the system 14
boundary, it is recommended that a material flow diagram is produced and used to account for all of 15
the material flows, e.g. within the processing stage the live weight is defined and shall equate the sum 16
of the mass of the products produced. 17
Spatial system boundary and stages: The cradle-to-farmgate stage includes feed and animal 18
components. The LCA of feeds is covered in detail in an associated document (LEAP Animal Feed 19
Guidlines) and covers the cradle-to-animal-mouth stage for all feed sources (including raw materials, 20
inputs, production, harvesting, storage, loss and feeding). Feeds may be grown on-farm, animals may 21
graze/browse across a range of feed sources on land with multiple ownership, and/or a proportion of 22
29
the feeds may be produced off-farm and transported to the farm for feeding to animals. The LEAP 1
Animal Feed Guidelines cover all emissions associated with land use and land-use-change. 2
The animal components cover all other inputs and emissions in the small ruminant supply chain not 3
covered by the LEAP Animal Feed Guidelines. This includes emissions associated with small 4
ruminant production and management. The latter includes accounting for the fate of excreta; however, 5
it is important to avoid double counting if excreta is captured as manure and represents a direct input 6
for feed production (manure emissions from transport and application are included in the LEAP 7
Animal Feed Guidelines). Animal production may involve more than one farm if animals are traded 8
between farms prior to going to processing (e.g. lambs/kids may be weaned or partly grown on one 9
farm and sold on to another farm for finishing). These multiple components shall be accounted for in 10
the calculations. 11
For the meat processing stage, there shall be no differentiation between the various products that are 12
edible by humans therefore impacts are divided evenly by mass over all such products from the small 13
ruminant supply chain because there are no significant biophysical or nutritional differences between 14
them (Section 11). 15
The primary processing stage is limited to animal slaughter (backyard, village slaughter unit and 16
abattoir) for meat processing to produce the functional unit. For primary processing in developing 17
countries, village slaughter centres are common. These can include direct processing as well as sale of 18
live animals to consumers for home processing or on-selling to large abattoirs near the cities. All 19
emissions directly related to inputs and activities in the cradle-to-primary processing chain stages are 20
included, irrespective of their location. All transportation steps within and between the cradle-to-21
primary processing gate are included. 22
The system boundaries covered shall include the feed production, animal production and primary 23
processing stages. 24
25
8.4.3 MATERIAL CONTRIBUTION AND THRESHOLD 26
In determining whether to expend resources and effort to include specific inputs, a 1 percent cut off 27
threshold for mass and energy should be adopted; larger thresholds shall be explicitly documented and 28
justified by the project goal and scope definition. Inputs to the system that represent less than 1 percent 29
of the mass or less than 1 percent of the energy required for a specific unit process (activity) in the 30
system can be excluded from the analysis. Instead, a scoping analysis (Section 8.2) can be used to 31
provide an estimate. An exception to this exclusion is made in cases where significant environmental 32
impact is associated with a small mass input (e.g. some material may be present in small quantities, yet 33
still have a relatively large environmental impact; these should be included). A minimum of 95 percent 34
of the impact for each category shall be accounted for. 35
30
8.4.4 TIME BOUNDARY FOR DATA 1
Documentation for temporal system boundaries shall describe how the assessment deviates from the 2
one-year time frame. By how many years is the temporal scope extended for the model parameters? 3
The time boundary for data shall be representative of the time period associated with the average GHG 4
emissions for the products. For products from small ruminants, a period of 12 months shall be used, 5
since this length of time is typical for the breeding cycle of sheep and goats. 6
In some cases where there may be considerable inter-annual variability in inputs, production and 7
emissions, it may be necessary for the one-year time boundary to be determined using data averaged 8
over 3 years to meet criteria of representativeness. An averaging period of 3 to 5 years is commonly 9
used to smooth the impact of seasonal and market variability on agricultural products. 10
11
8.4.5 CAPITAL GOODS 12
The production of capital goods (buildings and machinery) with a lifetime greater than one year may 13
be excluded in the life cycle inventory; however, the operation or occupation of, or other activities 14
utilizing capital goods shall be accounted. However, for studies in which the goal and scope include 15
assessment of alternate systems for which there may be significant differences in infrastructure 16
requirements, capital goods production shall be included. 17
18
8.4.6 ANCILLARY ACTIVITIES 19
Emissions from ancillary inputs, e.g. servicing, employee’s commutes, executive air travel or 20
accounting or legal services may be included if relevant. To determine if these activities are relevant, 21
an input-output analysis can be used as a scoping analysis. 22
23
8.4.7 DELAYED EMISSIONS 24
All emissions associated with products to the primary processing stage are assumed to occur within 25
the time boundary for data, generally of one year (Section 8.4.4). Delayed emissions from soil and 26
vegetation may be considered in the LEAP Animal Feel Guidelines. The PAS 2050:2011 provides 27
additional guidance regarding delayed emissions calculations for interested practitioners (BSI, 2011). 28
29
8.4.8 CARBON OFFSETS 30
Offsets shall not be included in the carbon footprint. However they may be reported separately as 31
“additional information”. 32
31
8.5 Impact categories 1
These guidelines are primarily based on assessment of GHG emissions. The total GHG emissions for 2
individual gases are summed along the system boundary. Individual gases are then multiplied by the 3
relevant characterization factor to convert them all into a common unit of carbon dioxide equivalents 4
(kg CO2eq). The characterization factors shall be based on the Global Warming Potentials of the 5
specific gases over a 100-year time horizon using the most recent IPCC factors which can be found in 6
the latest IPCC guidance documentation. Because characterization factors change as our understanding 7
evolves, it is important to note in the report documentation what specific sources were used for them. 8
The fossil energy demand should also be calculated, since all inputs of fossil fuels have to be 9
determined as part of the data collection requirements for assessing GHG emissions. This is captured 10
in the impact category called Cumulative Energy Demand and sub-category of non-renewable energy 11
resources, and uses the Lower Heating Value of the fuel for its characterization factor (Huijbregts et 12
al., 2006). It shall account for the embodied primary energy for the production and combustion of the 13
various energy sources and may draw on recognised databases such as Ecoinvent (Frischknecht et al., 14
2005). Fossil energy demand for the production and use of electricity, which will be specific for a 15
particular country, shall also be included. 16
The LCA of products should account for a range of resource use and environmental impact categories. 17
It is intended that in future these guidelines be updated to include multiple categories (Section 5.3). 18
19
9 MULTI‐FUNCTIONAL PROCESSES AND ALLOCATION 20
One of the challenges in LCA has always been associated with proper assignment (allocation) of 21
shared inputs and emissions to the multiple products from multi-functional processes. The choice of 22
the method for handling co-production often has a significant impact on the final distribution of 23
impacts across the co-products. Whichever procedure is adopted shall be documented and explained, 24
including sensitivity analysis of the choice on the results. As far as feasible, multi-functional 25
procedures should be applied consistently within and among the data sets. For situations where system 26
separation or expansion is not used, the allocated inputs and outputs should equal unallocated inputs 27
and outputs. 28
29
9.1 General principles 30
The ISO 14044 standard gives the following guidelines for LCA practitioners with respect to practices 31
for handling multi-functional production: 32
32
Step 1: Wherever possible, allocation should be avoided by: 1
a) dividing the unit process to be allocated into two or more sub-processes and collecting the 2
input and output data related to these sub-processes; or 3
b) expanding the product system to include the additional functions related to the co-products. 4
Step 2: Where allocation cannot be avoided, the inputs and outputs of the system should be partitioned 5
between its different products or functions in a way that reflects the underlying physical relationships 6
between them: i.e. they should reflect the way in which the inputs and outputs are changed by 7
quantitative changes in the products or functions delivered by the system. 8
Step 3: Where physical relationship alone cannot be established or used as the basis for allocation, the 9
inputs should be allocated between the products and functions in a way that reflects other relationships 10
between them. For example, input and output data might be allocated between co-products in 11
proportion to their economic value. 12
Where allocation of inputs is required, for example allocation of process energy between small 13
ruminant meat and other non-human edible products, the allocation procedures should follow the ISO 14
14044 allocation hierarchy. When allocation choices significantly affect the results, a sensitivity 15
analysis shall be performed to ensure robustness of conclusions. 16
Below is a list of commonly used procedures for addressing multi-functional processes, as stated, for 17
example, in ISO 14044 (2006, Section 9). Allocation can be based on other relationships such as: 18
bio-physical causality, arising from underlying biological or physical principles such as 19
material or energy balances; 20
physical properties such as mass, or protein or energy content; 21
economic value (revenue share) based on market prices of products. 22
Avoided burden or system expansion can be based on: 23
displacing average; 24
displacing marginal; 25
differentiating whether one is dealing with a determining or non-determining product flow, or 26
an avoided burden followed by sharing of credit. 27
28
9.2 A decision tree to guide methodology choices 29
A decision tree diagram to help decide on the appropriate methodology for dealing with co-products is 30
given in Figure 7. This involves a three-step approach and the principles involved in working through 31
it are as follows: 32
33
Step 1: Avoid allocation by subdividing the processing system into three categories. 1 A production unit is defined here as a group of activities (and the necessary inputs, machinery and 2
equipment) in a processing facility or a farm that are needed to produce one or more co-products. 3
Examples are the crop fields in an arable farm, the different animal herds (sheep, goats, cattle, deer), 4
or the individual processing lines in a manufacturing facility. 5
In the first step (ISO step (1a) subdivision) all processes and activities of a farm/processing facility are 6
divided into three categories: the condition is that inputs/emissions are easily divided over the 7
processes and that enough information about the inputs/emissions is available. 8
flow 1.a. Inputs/activities that should be directly assigned to a single co-product, e.g. packaging and 9
post-processing storage for meat products, or rendering energy requirements in the post-10
exsanguination phase at the processing plant. 11
flow 1.b. Inputs/activities that should be assigned to production units that can provide single or 12
multiple co-products, e.g. input of pesticides or energy inputs of operations of a barn or 13
manufacturing facility, or feed for a specific animal type at a farm. 14
flow 1.c. Inputs/activities of a nonspecific nature in a farm or processing facility such as heating, 15
ventilation, climate control, internal transport in a manufacturing facility or farm that 16
cannot be directly attributed to production units. For example energy to pump drinking 17
water for multiple animal species in a small-scale, multi-species operation. 18
34
FIGURE 7: MULTI‐FUNCTIONAL OUTPUT DECISION TREE 1
2 Note: The choice of method for handling multi-functional outputs for each stage or process in the supply chain shall be based on this decision algorithm. 3
35
Step 2. Attribute joint production to production units 1 In theory, all joint production systems are separable, where sufficiently detailed data exist, and should 2
normally follow path 1a. Nevertheless, situations exist where this is impractical, and in step 2 3
(Figure 7) the generic processes should be attributed to production units on the basis of ISO steps 1b, 2 4
and 3. For example sheep, goats, cattle, alpaca and deer may be all raised in a single production unit. 5
In this situation, farm overhead operations that cannot be explicitly assigned to an individual species 6
should be handled using the criteria in box/Step 2. For some production systems, the 1b path to box 3 7
will be followed, as the inputs and outputs in a single animal-species system are clearly assigned to the 8
single production unit and its activities/operations and products. 9
System expansion: ISO step (1b) should only be applied on condition that the avoided production 10
system can be unambiguously determined and there is little interference with other feed or animal 11
production systems (flow 2a in Figure 7). 12
If an output generated by the production unit unambiguously avoids external production, apply 13
substitution (system expansion). There is one important condition for system expansion: that the 14
avoided co-product shall be fully equivalent. This means that you shall be very certain and clear about 15
the avoided production and that there should be no, or minimal interference with other production 16
systems. The result of this step will be an avoided product in the inventory for the unit process; 17
however, at the inventory level there is no corresponding reduction in the emissions or exchanges with 18
the environment. The credit for the substitution will arise in subsequent calculations. 19
When system expansion is not possible, the second question is whether a physical allocation is 20
possible. The condition here is that the products should have similar physical properties and serve 21
similar goals or markets, or that known processing or biophysical relationships can be used to assign 22
inputs and outputs of a single production unit to each product. For example, if feed is provided to 23
multiple animal species, the animal growth requirements may be used to apportion the shared feed 24
between the species. The result of this step will be a splitting of some inventory flows between the 25
production units, and if the resultant unit process is multifunctional, these inventory flows will be 26
allocated to single co-products in the next step of the procedure (box 3 in Figure 7). 27
If inputs in a multiple production system benefit all products and cannot be specifically assigned to 28
production units, the allocation should be preferably based on a physical property or mechanistic 29
algorithm (flow 2b in Figure 7). 30
When physical allocation is not possible or allowed, the last option is economic allocation. As with 31
physical allocation, the result of this step will be a splitting of some inventory flows between the 32
production units, and if the resultant unit process is multifunctional, these inventory flows will be 33
allocated to single co-products in the next step of the procedure (Box 3 in Figure 7). 34
36
Step 3. Split single production units into individual co-products 1 After performing steps 1 and 2, all inputs and operations will have been attributed to the single 2
production unit, or already to a single product. An inventory table is made for the production unit. 3
Step 3 guides the assignment of inputs and emissions from a single production unit to each co-product 4
produced by the unit. If there is only a single product at this stage, the process is complete. In the case 5
of a multifunctional production unit, the first approach is to employ system expansion. The same rule 6
holds as the one defined above for production units, so only in very clear situations of avoidance, such 7
as connection to an electricity grid, system expansion should be applied. Any flow arising from 3a will 8
follow this path. When system expansion is not possible, the remaining outputs must be classified as 9
co-products, residual products or wastes. Existing standards state that a footprint should not get a 10
credit for sold electricity (or natural gas) that is incorporated into grid factors (WRI 2011; BSI 2011). 11
Therefore it is not valid, according to these existing standards, to provide a credit using system 12
expansion (or other method) for sold electricity/gas. Further, system expansion does not provide a 13
footprint for the avoided product that can be used in another LCA (for a downstream system). This 14
guidance strongly encourages other methods for allocation, in particular when there is a danger for 15
improper interpretation at policy level when several LCAs are combined to obtain an aggregated view 16
of the larger system; specifically, if an animal production system uses system expansion, and a later 17
crop production analysis assumes the credit for use of manure, there will be a ‘double credit’ in the 18
combined system. 19
Outputs of a production process are considered as residual if (flow 3f): 20
they are sold in the condition in which they appear in the process and contribute very little to 21
the turnover of the company (total value of the flow less than 1 percent); 22
the upstream and production process that produce the outputs are not deliberately modified for 23
the outputs, and has a subsequent use. There may be value-added steps beyond the boundary 24
of the small ruminant system under study, but these activities do not impact the small 25
ruminant system calculations in these guidelines. 26
Residual products will not receive any allocated emissions, nor will they contribute emissions to the 27
main co-products of the production unit. However it is useful to track residual flows for the purpose of 28
understanding the mass balance for the production unit. 29
An output of a production process shall be considered as waste if it carries no economic value or if the 30
production unit incurs a cost for treatment or removal. Waste has to be treated and/or disposed of and 31
these emissions shall be included in the inventory. For the small ruminant sector, the most common 32
process falling in this category is wastewater treatment at processing facilities and, in some cases, 33
manure sent to a landfill. Manure exported off-farm is discussed below. 34
37
All other co-products are subject to allocation, leading to flows 3b, 3c and 3d in Figure 7. Assignment 1
to these flows depends upon whether physical, biophysical or mechanistic allocation is possible (3b), 2
or whether an economic allocation at a single product (3c) or product group level (3d) is applied. 3
The condition for determining whether physical allocation is appropriate is that the products should 4
have similar physical properties and serve similar functions or markets, or a straightforward 5
mechanistic algorithm can be identified, or that biophysical relationships can be used to assign inputs 6
and emissions to each product. When physical allocation is not feasible (interactions too complex to 7
accurately define a mechanistic relationship) or is not allowed (dissimilar properties or markets), the 8
last option is economic allocation. 9
In the case of economic allocation, one option (flow 3d) is grouping a number of co-products and 10
performing the allocation with some co-products at the group level instead of the single product level. 11
This option is relevant for the various edible meat components (e.g. carcass cuts and edible offal) 12
which can be grouped before allocation between them and other inedible co-products such as hide, 13
blood and renderables. 14
15
9.2.1 ALLOCATION OF TRANSPORT 16
Estimating environmental impacts of transportation entails two allocation issues: allocation of empty 17
transport distance covered by transport means and allocation of the load fraction of transportation 18
means. The allocation of empty transport distance is often already included in the models used when 19
the secondary life cycle inventory (LCI) data for transportation were created. However if primary data 20
for transport are derived, the LCA practitioner shall make an estimate of the empty transport distance. 21
In the absence of other information, a worst-case estimate should be applied here, meaning inclusion 22
of 100% extra transport for empty return. Allocation of empty transport kilometres shall be done on 23
the basis of the average load factor for the transport under study. If no supporting information is 24
collected, it shall be assumed that 100% extra transport is needed for empty return. 25
If products are transported by a vehicle, resource use and emissions of the vehicle shall be allocated to 26
the transported products. A means of transport has a maximum load. This maximum load is expressed 27
in tonnes. However the maximum weight can only be achieved if density of the loaded goods allows 28
it. Therefore allocation of transport emissions to transported products shall be done on the basis of 29
mass share unless the density of the transported product is significantly lower than average so that the 30
volume restricts the maximum load. 31
38
9.3 Application of general principles for small ruminant systems and processes 1
In practice, dealing with multi-functional processes and the choice of allocation method is a 2
contentious issue in LCA studies and for small ruminants there are a number of steps where allocation 3
decisions are required. Thus, this guidelines document goes into some detail on each of these steps and 4
gives recommendations on the preferred allocation methodology for each one (Section 11.2.5 and 5
11.6). The recommended methods, based on use of the decision tree, are summarized in Table 2. 6
Cradle-to-farm-gate: Within the cradle-to-farm-gate stage there are a number of allocation decisions 7
associated with feeds; these are described in the LEAP Animal Feed Guidelines. Within the animal 8
production stage, there are two main areas where co-products need to be accounted for. These are: 9
where different animal species consume the same feed source(s) and/or share non-feed related 10
inputs; 11
where small ruminants produce multiple products of meat, milk and fibre. 12
In ruminant livestock systems, the major determinant of GHG emissions is enteric methane and 13
excreta N2O emissions, and the driver of these is feed intake. Consequently, if the activities, inputs or 14
emissions cannot be separated, the preferred method to account for multi-functional processes and co-15
products shall be a biophysical approach based on feed intake associated with the different animal 16
species or co-products. 17
In practice, accounting for multiple animal species (i.e. step 1c in Figure 7, since this is not a single 18
production unit) is based firstly on separation of activities between species and then on determination 19
of feed intake for each species (i.e. step 2b in Figure 7). Remaining shared inputs (e.g. energy use for 20
water provision and animal movement) are allocated according to relative feed intake between species. 21
At a farm level, the equivalent output from this approach would be to determine all feed and animal 22
related emissions for the farm and use the allocation value for the target small ruminant species based 23
on relative feed intake to determine that species’ total emissions. 24
39
TABLE 2: RECOMMENDED METHODS FOR DEALING WITH MULTI‐FUNCTIONAL PROCESSES AND ALLOCATION 1 BETWEEN CO‐PRODUCTS FOR THE CRADLE‐TO‐PRIMARY PROCESSING GATE STAGES OF THE LIFE CYCLE OF SMALL 2 RUMINANT PRODUCTS 3
Source/stage
of co‐products
Recommended
method* Basis
Animal species
(within farm)
1. Separate farm activities
2. Biophysical causality
First, separate the activities specific to an animal species. Then, determine
emissions specific to feeds relating to the sheep or goats under study. For
remaining non‐feed inputs, use biophysical allocation based on the
proportion of total energy requirements for each of the different animal
species.
Meat, fibre, milk
(within farm)
1. Separate activities
2. Biophysical causality
First, separate activities specific to products (e.g. electricity for shearing or
milking). Then use biophysical allocation according to energy or protein
requirements for animal physiological functions of growth, fibre production,
milk production, reproduction and maintenance.
Milk processing to
milk products
1. Separate activities
2. Mass of fat + protein
First, separate activities specific to individual products where possible. Then
use allocation based on the relative amount of fat + protein in the milk
products
Fibre processing to
clean fibre and
lanolin
1. Separate activities
2. Economic
First, separate the activities specific to individual products where possible.
Then use economic allocation based on a minimum of three years of recent
average prices.
Meat processing to
meat and non‐
meat products
1. Separate activities
2. Economic
First, separate the activities specific to individual products where possible.
Then use economic allocation based a minimum of three years of recent
average prices.
Note: * Where choice of allocation can have a significant effect on results, it is recommended to use more than one method to 4 illustrate the effects of choice of allocation methodology. 5 6
For sheep or goats (i.e. which are a single production unit and therefore follow step 1b), the allocation 7
between meat, milk and fibre co-products shall be based on biophysical allocation according to feed 8
requirements for production of the products (described in detail in Section 11.2.5; steps 3a1 and 3b in 9
Figure 7). This aligns with the IDF (2010a) methodology for allocation between milk and meat for 10
dairy cows. Previous studies have shown that the choice of allocation method for meat, milk and fibre 11
co-products can have a significant effect on reported GHG emissions (e.g. Ledgard et al., 2011; Gac 12
et al., 2012). As noted previously, where choice of allocation can have a significant effect on results, 13
more than one method shall be used to illustrate the effects of choice of allocation methodology. 14
Alternate methodological approaches include: system expansion, economic allocation and mass 15
allocation. This is also important when the guidelines are used for analysing the implications for co-16
products and the potential benefits of mitigation options. For example, depending on the methodology 17
employed, the use of mitigation to reduce emissions from a main product may have unintended effects 18
on increasing emissions from co-products and their associated production systems, leading to no 19
overall benefits (e.g. Flysjö et al., 2012). 20
40
Manure exported off-farm: In some situations, part or all of the excreta from animals may be 1
collected and sent off-farm. In practice, this is of minor importance in most small ruminant systems 2
and detailed discussion on options for handling manure exported off-farm are given in Appendix 4. It 3
is the strong recommendation of these guidelines that the default method for manure should be to 4
consider it as a residual at the farm gate. This is most applicable when manure has essentially no value 5
as it leaves the farm gate. In such cases, emissions linked to the production and storage on-farm of 6
manure would go to the other co-product(s) of live-weight, milk and greasy fibre. Appendix 4 outlines 7
other options including system expansion relating to energy or fertilizer, or treating manure as a co-8
product, residual or waste. 9
Primary processing: For the milk primary processing stage (a single production unit following steps 1b, 10
3a1 and 3b in Figure 7), allocation between co-products shall be based on the relative mass of fat + 11
protein. The use of this approach for dairy products aligns with that used in recent publications for milk 12
products from dairy cows (Flysjö et al., 2011; Thoma et al., 2013b), and meets the requirements for 13
similarity of products in Figure 7. However, for fibre and meat-processing systems the products have 14
very different end uses (except in meat processing for offal and meat cuts, which are seen as having the 15
same function and are considered together in the same product group of “edible meat products”); 16
therefore, economic allocation is considered the most appropriate approach using Figure 7 (i.e. steps 3b1 17
and 3c or 3d). It is recognized that some co-products may be identified as being of no economic value 18
after primary processing (and in Figure 7 would be classified as a residual; step 3f), but that they may be 19
collected and used for secondary processing (e.g. used for burning for energy or used for producing 20
blood-and-bone meal). However, in that case the product, having undergone secondary processing, is 21
considered to fall outside the system boundary for these guidelines. 22
23
10 COMPILING AND RECORDING INVENTORY DATA 24
10.1 General principles 25
The compilation of the inventory data should be aligned with the goal and scope of the life cycle 26
assessment. The LEAP guidelines are intended to provide LCA practitioners with practical advice for 27
a range of potential study objectives. This is in recognition of the fact that studies may wish to assess 28
small ruminant supply chains ranging from individual farms, to integrated production systems, to 29
regional or national scale, or to a sector level. When evaluating the data collection requirements for the 30
project, it is necessary to consider the influence of the project scope. In general these guidelines 31
recommend collection of primary activity data (Section 10.2.1) for foreground processes, those 32
processes generally being considered as under the control or direct influence of the study 33
commissioner; however, it is recognized that for projects with larger scope, such as sectoral analyses 34
at the national scale, the collection of primary data for all foreground processes may be impractical. In 35
41
such situations, or when an LCA is conducted for policy analysis, foreground systems may be 1
modelled using data obtained from secondary sources such as national statistical databases, peer-2
reviewed literature or other reputable sources. 3
An inventory of all materials, energy resource inputs, outputs (including products, co-products and 4
emissions) for the product supply chain under study shall be compiled. The data recorded in relation to this 5
inventory shall include all processes and emissions occurring within the system boundary of that product. 6
As far as possible, primary inventory data shall be collected for all resource use and emissions 7
associated with each life cycle stage included in the defined system boundaries. For processes where 8
the practitioner does not have direct access to primary data (i.e. background processes), secondary data 9
can be used. Data collected directly from suppliers should be used for the most relevant products 10
supplied by them when possible. If secondary data are more representative or appropriate than primary 11
data for foreground processes (to be justified and reported), secondary data shall also be used for these 12
foreground processes. 13
For agricultural systems, two main differences exist compared to industrial systems. Firstly, 14
production may not be static from year to year, and secondly, some inputs and outputs are very 15
difficult to measure. Consequently, the inventory stage of an agricultural LCA is far more complex 16
than most industrial processes, and may require extensive modelling in order to define the inputs and 17
outputs from the system. For this reason agricultural studies often rely on a far smaller sample size 18
and are often presented as ‘case studies’ rather than ‘industry averages’. For agricultural systems, 19
many foreground processes must be modelled or estimated rather than measured. Assumptions made 20
during the inventory development are critical to the results of the study and need to be carefully 21
explained in the methodology of the study. In order to clarify the nature of the inventory data, it is 22
useful to differentiate between ‘measured’ and ‘modelled’ foreground system LCI data. 23
The LCA practitioner shall demonstrate that the following aspects in data collection have been taken 24
into consideration when carrying out the assessment (adapted from ISO14044): 25
Representativeness: qualitative assessment of the degree to which the data set reflects the 26
true population of interest. Representativeness covers the three following dimensions: 27
a) time-related representativeness: age of data and the length of time over which data was 28
collected; 29
b) geographical representativeness: geographical area from which data for unit processes was 30
collected to satisfy the goal of the study; 31
c) technology representativeness: specific technology or technology mix; 32
Precision: measure of the variability of the data values for each data expressed (e.g. standard 33
deviation); 34
Completeness: percentage of flow that is measured or estimated; 35
42
Consistency: qualitative assessment of whether the study methodology is applied uniformly to 1
the various components of the analysis; 2
Reproducibility: qualitative assessment of the extent to which information about the 3
methodology and data values would allow an independent practitioner to reproduce the results 4
reported in the study; 5
Sources of the data; 6
Uncertainty of the information (e.g. data, models and assumptions). 7
For significant processes, LCA practitioner shall document the data sources, the data quality, and any 8
efforts made to improve data quality. 9
10
10.2 Requirements and guidance for the collection of data 11
Two types of data may be collected and used in performing LCAs: 12
Primary data: defined as directly measured or collected data representative of processes at a 13
specific facility or for specific processes within the product supply chain. 14
Secondary data: defined as information obtained from sources other than direct measurement 15
of the inputs/outputs (or purchases and emissions) from processes included in the life cycle of 16
the product (PAS 2050:2011, 3.41). Secondary data are used when primary data are not 17
available or it is impractical to obtain them. Some emissions, enteric fermentation in the 18
rumen of animals, are calculated from a model, and are therefore considered secondary data. 19
For projects where significant primary data is to be collected, a data management plan is a valuable 20
tool for managing data and tracking the process of LCI data set creation, including metadata 21
documentation. The data management plan should include (WRI 2011, Appendix C): 22
description of data collection procedures; 23
data sources; 24
calculation methodologies; 25
data transmission, storage and backup procedures; and 26
quality control and review procedures for data collection, input and handling activities, data 27
documentation and emissions calculations. 28
The recommended hierarchy of criteria for acceptance of data is: 29
primary data collected as part of the project and that have a documented Quality Assessment 30 (Section 10.3); 31
data from previous projects that have a documented Quality Assessment; 32
data published in peer-reviewed journals or from generally accepted LCA databases that are 33 regarded as reliable sources of information; 34
data presented at conferences or otherwise publicly available (e.g., internet sources); 35
data from industrial studies or reports can be considered. 36
43
10.2.1 REQUIREMENTS AND GUIDANCE FOR THE COLLECTION OF PRIMARY DATA 1
In general, primary data shall, to the fullest extent feasible, be collected for all foreground processes 2
and for the main contributing sources to GHG emissions. Foreground processes, here defined as those 3
processes under the direct control of, or significantly influenced by, the study commissioner, are 4
depicted in Figure 6 under feed, water and animals. Raw material acquisition represents background 5
data. For most systems, the production of feed on-farm is fully integrated into the production system 6
and is therefore a foreground process, whereas brought-in feeds from off-farm can be considered a 7
background process. Some foreground processes are impractical to measure for an LCA, for example, 8
methane emissions from animal rumen fermentation and manure management. In cases such as this, 9
when a model is used to estimate the emission, the input data used for the model shall be measured. 10
The practicality of measured data for all foreground processes is also related to the scale of the project. 11
As an example, if a national-scale evaluation of the small ruminant sector is planned, it is impractical 12
to collect farm-level data from all small ruminant producers. In these cases, aggregated data from 13
national statistical databases or other sources (e.g., trade organizations) may be used for foreground 14
processes. In every case, clear documentation of the data collection process and data quality 15
documentation to ensure compatibility with the study goal and scope shall be incorporated into the 16
report. 17
Relevant specific data shall be collected that is representative for the product or processes being 18
assessed. To the greatest extent possible, recent data shall be used, such as current data from industry 19
stakeholders. Data shall be collected that respect geographic relevance (e.g. for crop yield in relation 20
to climate and soils) and align to the defined goal and scope of the analysis. Each data source should 21
be acknowledged and uncertainty in the data quality noted. 22
Prior work (e.g. Appendix 1) has identified the main hotspots and primary data (or modelled estimates 23
using primary input data) shall be used for these stages of the supply chain. Specifically, the cradle-to-24
farm-gate stage can dominate whole life cycle emissions (e.g. c. 80 percent in Ledgard et al., 2011) 25
and animal enteric methane can represent around 50-70 percent of cradle-to-farm-gate emissions. 26
Thus, animal population and productivity, and feed quality data are key primary activity data needed 27
to calculate enteric methane emissions. Similarly, methane and N2O from animal excreta can represent 28
about 5-35 percent of cradle-to-farm-gate emissions and also require data on feed composition and 29
chemical analysis to calculate them. Where manure is collected from animals, methods of storage and 30
use can have a significant effect on emissions. Primary activity data on this area is therefore required. 31
The contribution from emissions associated with feed production can vary greatly from minimal in 32
low-input extensive grassland/rangeland/nomadic/transhumance systems to about 40 percent in 33
intensive crop-based or zero-grazing systems. Corresponding direct on-farm energy use is also 34
variable from minimal to about 20 percent, with a global average of about 2 percent (Gerber et al., 35
2013). In a whole life cycle analysis, Ledgard et al. (2011) showed that transportation accounted for 36
44
up to 5 percent of total GHG emissions, including shipping of product up to 20,000 km to distant 1
markets. The study showed that all emissions associated with meat processing (abattoirs) represented 2
3 percent of total life cycle emissions, of which fuel use, electricity use and wastewater processing 3
were the dominant contributors. 4
5
10.2.2 REQUIREMENTS AND GUIDANCE FOR THE COLLECTION AND USE OF SECONDARY DATA 6
Secondary data refers to life cycle inventory data sets generally available from existing third-party 7
databases, government or industry association reports, peer-reviewed literature, or other sources. It is 8
normally used for background system processes, such as electricity or diesel fuel which may be 9
consumed by foreground system processes. When using secondary data it is necessary to selectively 10
choose the data sets which will be incorporated into the analysis. Specifically, life cycle inventory for 11
goods and services consumed by the foreground system should be geographically and technically 12
relevant. An assessment of the quality of these data sets (Section 10.3.2) for use in the specific 13
application should be made and included in the documentation of the data quality analysis. 14
Where primary data are unavailable and where inputs or processes make a minor contribution to total 15
GHG emissions, secondary or default data may be used. However, geographic relevance should be 16
considered. For example, if default data are used for a minor input, such as a pesticide, the source of 17
production should be determined and a transportation component added to calculation of the emissions 18
to account for its delivery from site of production to site of use. Similarly, where there is an electricity 19
component related to an input, a relevant electricity emission factor for the country or site of use 20
should be used that accounts for the relevant energy grid mix. 21
Secondary data should only be used for foreground processes if primary data are unavailable, if the 22
process is not environmentally significant, or if the goal and scope permit secondary data from 23
national databases or equivalent sources. All secondary data should satisfy the following requirements: 24
They shall be as current as possible and collected within the past 5-7 years. 25
They should be used only for processes in the background system. When available, sector-26
specific data shall be used instead of proxy LCI data. 27
They shall fulfill the data quality requirements specified in this guide (Section 10.3). 28
They should, where available, be sourced following the data sources provided in this guide 29
(e.g. Section 11.2.2 for animal assessment and in Appendices 3 and 5). 30
They may only be used for foreground processes if specific data are unavailable or the process 31
is not environmentally significant. However, if the quality of available specific data is 32
considerably lower and the proxy or average data sufficiently represents the process, then 33
proxy data shall be used. 34
45
An assessment of the quality of these data sets for use in the specific application should be made and 1
included in the documentation of the data quality analysis. 2
3
10.2.3 APPROACHES FOR ADDRESSING DATA GAPS IN LCI 4
Data gaps exist when there is no primary or secondary data available that is sufficiently representative 5
of the given process in the product’s life cycle. LCI data gaps can result in inaccurate and erroneous 6
results (Reap et al., 2008). When missing LCI is set to zero, the result is bias towards lower 7
environmental impacts (Huijbregts et al., 2001). 8
Several approaches have been used to bridge data gaps, but none are considered standard LCA 9
methodology (Finnveden et al. 2009). As much as possible, the LCA practitioner shall attempt to fill 10
data gaps by collecting the missing data. However, data collection is time-consuming and expensive, 11
and is often not feasible. The following sections provide additional guidance on filling data gaps with 12
proxy and estimated data. 13
The use of proxy data sets – background LCI data sets which are the most similar process/product for 14
which data is available – is common. This technique relies on the practitioner's judgment, and is 15
therefore, at least arguably, arbitrary (Huijbregts et al. 2001). Using the average of several proxy data 16
sets has been suggested as a means to reduce uncertainty compared to the use of a single data set (Mila 17
i Canals et al. 2011). Mila i Canals et al. 2011 also suggest that extrapolation from one data set to 18
bridge the gap may also be used. For example, data from goat and sheep production could be 19
extrapolated for production of other small ruminants (e.g. deer, llama, alpaca), based on expert 20
knowledge of differences in feed requirements, feed conversion ratios, excreta characteristics, and 21
fibre or antler production. Adapting an energy emission factor for one region to another with a 22
different generation mix is another example. While use of proxy datasets is the simplest solution, it 23
also has the highest element of uncertainty. Extrapolation methods require expert knowledge and are 24
more difficult to apply, but provide more accurate results. 25
For countries where environmentally extended economic input-output tables have been produced, a 26
hybrid approach can also be used as a means of bridging data gaps. In this approach the monitor value 27
of the missing input is analysed through the input-output tables and then used as a proxy LCI data set. 28
This approach is of course subject to uncertainty and has been criticized (Finnveden et al., 2009). 29
Any data gaps shall be filled using the best available secondary or extrapolated data. The contribution 30
of such data (including gaps in secondary data) shall not account for more than 20 percent of the 31
overall contribution to each emission factor impact category considered. 32
In line with the guidance on data quality assessment, any assumptions made in filling data gaps, along 33
with the anticipated effect on the product inventory final results, shall be documented. If possible, the 34
46
use of such gap-filling data should be accompanied by data quality indicators, such as a range of 1
values or statistical measures that convey information about the possible error associated with using 2
the chosen method. 3
4
10.3 Data quality assessment 5
LCA practitioners are required to assess data quality by using data quality indicators. Generally, data 6
quality assessment can indicate how representative the data are as well as their quality. Assessing data 7
quality is important for a number of reasons: improving the inventory’s data content, for proper 8
communication and interpretation of results, as well as informing users about the possible uses of the 9
data. Data quality refers to characteristics of data that relate to their ability to satisfy stated 10
requirements (ISO14040: 2006). Data quality covers various aspects, such as technological, 11
geographical and time-related-representativeness, as well as completeness and precision of the 12
inventory data. This section describes how the data quality shall be assessed. 13
14
10.3.1 DATA QUALITY RULES 15
Criteria for assessing LCI data quality can be structured by representativeness (technological, 16
geographical, and time-related), completeness (regarding impact category coverage in the inventory), 17
precision/uncertainty (of the collected or modelled inventory data), and methodological 18
appropriateness and consistency. Representativeness addresses how well the collected inventory data 19
represents the “true” inventory of the process for which they are collected regarding technology, 20
geography and time. For data quality, the representativeness of the LCI data is a key component and 21
primary data gathered shall adhere to the data quality criteria of technological, geographical, and time-22
related representativeness. Table 3 presents a summary of requirements for data quality. Any 23
deviations from the requirements shall be documented. Data quality requirements shall apply to both 24
primary and secondary data. For LCA studies using actual farm data and targeted at addressing farmer 25
behaviour, ensuring that farms surveyed are representative and the data collected is of good quality 26
and well managed is more important than detailed uncertainty assessment. 27
47
TABLE 3: OVERVIEW OF REQUIREMENTS FOR DATA QUALITY 1
2
10.3.2 DATA QUALITY INDICATORS 3
Data quality indicators define the standard for the data to be collected. These standards relate to issues 4
such as representativeness, age, system boundaries, etc. During the data collection process, data 5
quality of activity data, emission factors, and/ or direct emissions data shall be assessed using the data 6
quality indicators. WRI/WBCSD has published additional guidance on quantitative uncertainty 7
assessment3 which includes a spreadsheet to assist in the calculations. 8
Data collected from primary sources should be checked for validity by ensuring consistency of units 9
for reporting and conversion as well as material balances to ensure that, for example, all incoming 10
materials are accounted in products leaving the processing facility. 11
Secondary data for background processes can be obtained from, for example, the EcoInvent database. 12
In this situation, the data quality information provided by the database manager should be evaluated to 13
determine if it requires modification for the study underway – for example, if use of European 14
electricity grid processes in other areas will increase the uncertainty of those unit processes. 15
16
10.4 Uncertainty analysis and related data collection 17
Data with high uncertainty can negatively impact the overall quality of the inventory. The collection of 18
data for the uncertainty assessment and understanding uncertainty is crucial for the proper 19
interpretation of results (Section 12) and reporting and communication (Section 12.5). 20
21
3 http://www.ghgprotocol.org/calculation-tools/all-tools
Indicator Requirements/ data quality rules
Technological representativeness
The data gathered shall represent the processes under consideration.
Geographical representativeness:
If multiple units are under consideration for the collection of primary data, the data gathered shall, at a minimum, represent a local region such as EU‐27.
Data should be collected respecting geographic relevance to the defined goal and scope of the analysis.
Temporal representativeness Primary data gathered shall be representative for the past three years and 5‐7 years for secondary data sources.
The representative time period on which data is based shall be documented.
48
The following guidelines shall apply for all studies intended for distribution to third parties, and 1
should be followed for internal studies intended for process improvement: 2
Whenever data is gathered, data should also be collected for the uncertainty assessment. 3
Gathered data should be presented as a best estimate or average value, with an uncertainty 4
indication in the form a standard deviation (where plus and minus twice the standard deviation 5
indicates the 95% confidence interval if data follow a normal distribution. 6
When a large set of data is available, the standard deviation should be calculated directly from 7
this data. For single data points, the bandwidth shall be estimated. In both cases the 8
calculations or assumptions for estimates shall be documented. 9
10
10.4.1 SECONDARY ACTIVITY DATA 11
See Section 10.2.2. 12
13
10.4.2 DEFAULT/PROXY DATA 14
See Section 10.2.2. 15
16
10.4.3 INTER‐ AND INTRA‐ANNUAL VARIABILITY IN EMISSIONS 17
Agricultural processes are highly susceptible to variations in weather patterns year-to-year. This is 18
particularly true for crop yields, but may also affect feed conversion ratios when environmental 19
conditions are severe enough to have an impact on an animal’s performance. Depending on the goal 20
and scope definition for the study, additional information may be warranted such that either seasonal 21
or inter-annual variability in the product system efficiency can be captured and identified. 22
23
10.4.4 INTER‐ AND INTRA‐ANNUAL VARIABILITY IN EMISSIONS 24
Agricultural processes are highly susceptible to variations in weather patterns year-to-year. This is 25
particularly true for crop yields, but may also affect feed conversion ratios when the environmental 26
conditions are severe enough to have an impact on an animal’s performance. Depending on the goal 27
and scope definition for the study, additional information may be warranted such that either seasonal 28
for inter-annual variability in the product system efficiency can be captured and identified. 29
49
11 LIFE CYCLE INVENTORY 1
11.1 Overview 2
The life cycle inventory (LCI) analysis phase involves the collection and quantification of inputs and 3
outputs throughout the life cycle stages covered by the system boundary of the study (Figure 6). This 4
typically involves an iterative process (as described in ISO 14044: 2006), with the first steps involving 5
data collection using principles as outlined in Section 10. The subsequent steps in this process involve 6
recording and validation of the data; relating the data to each unit process and functional unit 7
(including allocation for different co-products); and aggregation of data, ensuring all significant 8
processes, inputs and outputs are included within the system boundary. 9
The system boundary (Figure 6) has pre and post farm gate stages. 10
11
11.2 Cradletofarmgate 12
The cradle-to-farm-gate stage consists of three main processes of raw material acquisition, water and 13
feed production, and use for animal production (Figure 8). Most raw material acquisition is associated 14
with the production of feeds. Note that these guidelines only provide limited background information 15
related to animal feeds because these are covered in the separate LEAP Animal Feed Guidelines 16
document. Thus, animal feeds information is presented largely for context, but also because of the 17
strong linkage between feeds and animal production. 18
50
FIGURE 8: PROCESSES THAT CONTRIBUTE TO GHG EMISSIONS AND FOSSIL ENERGY DEMAND COVERING RAW 1 MATERIALS, WATER USE, FEED PRODUCTION AND USE, AND ANIMAL PRODUCTION WITHIN THE SYSTEM BOUNDARY 2 OF THE CRADLE‐TO‐FARM‐GATE 3
4
Note: The box with a blue background refers to inputs, processes and emissions covered by the LEAP Animal Feed Guidelines 5 and not part of the current guidelines. 6 7
Water supply to animals is important for their survival and energy inputs are often required for the 8
provision (e.g. for pumping and reticulation) and/or transport of water. The GHG emissions associated 9
with this and other sources of energy use shall be included. There is also a small contribution to 10
resource use and GHG emissions associated with the production and provision of animal health inputs, 11
which may include treatments for infectious diseases, internal and external parasites, reproductive 12
diseases and metabolic diseases (e.g. Besier et al., 2010). 13
To assist the user in working through the process of calculating the carbon footprint of products for the 14
cradle-to-farm-gate stage, a flow diagram illustrating the various steps involved is presented in 15
Figure 9. 16
At the cradle-to-farm-gate stage, previous research has shown that the largest single source of GHG 17
emissions is methane from digestion of feeds in the rumen of goats and sheep. For example, Ledgard 18
51
et al. (2011) estimated enteric methane at 57 percent of the total life cycle emissions for lamb from 1
New Zealand consumed in England. Thus, it is important to obtain an accurate estimate of feed intake 2
by small ruminants. This aspect is covered in detail in Section 11.2.2. However, an important first step 3
is to define the feed types used and their feed quality characteristics. 4
5
11.2.1 FEED ASSESSMENT 6
The production, conservation and use of feeds can represent a significant contributor to the total 7
resource use and GHG emissions from small ruminant products. Thus, it is important to accurately 8
identify the number and types of feeds used, which can vary markedly in different small ruminant 9
production systems, as discussed in Section 6.2. Determination of the amount of each feed used is 10
described in detail in Section 11.2.2. 11
Feed types can vary from annual crops (where the feed source may be harvested grains, whole crop 12
silage/hay or forage crops grazed in situ) through to perennial plants. The latter includes pastures, 13
range forages, browse and tree cuttings. A summary of the typical composition (dry matter, energy, 14
protein, fibre and phosphorus concentrations) of a very wide range of these feed types is given in NRC 15
(2007; Tables 15.11 and 15.12). Primary data on the composition of the main feed sources used shall 16
be obtained for use in the LCI analysis wherever possible, but the NRC tables provide default values 17
when primary data cannot be obtained. 18
19
a) Calculating environmental impacts of feed production 20
The LEAP Guidelines document on the LCA of feeds describes the methodology for calculation of 21
GHG emissions associated with the production, processing and storage of animal feeds. The main raw 22
materials and processes that must be accounted for in determining the emissions of feeds are given in 23
Figure 8. Key contributors to GHG emissions are inputs of fertilizers, manures and lime (including 24
relevant manufacturing, transport and application stages); fuel used for production, processing and 25
transport; crop residues (that produce N2O emissions) and land use change. Land use change and 26
carbon sequestration in soil can be important contributors to GHG emissions or removals, but these 27
relate specifically to the feed production stage and, therefore, these aspects are covered in the LEAP 28
Animal Feed Guidelines (see also BSI, 2011 for methods). 29
A wide range of processed feeds or concentrates are used globally and various databases are being 30
developed by a number of groups (including FAO) that could provide default values for the total GHG 31
emissions per kg feed. Default values are appropriate where relevant country-specific data are 32
unavailable, and where their use is a minor component of the main feeds used. 33
52
FIGURE 9: FLOW DIAGRAM AS A GUIDE TO THE PROCEDURE FOR DETERMINING THE CARBON FOOTPRINT OF SMALL 1 RUMINANT PRODUCTS FOR THE CRADLE‐TO‐FARM‐GATE STAGE 2
3
Note: Content within the ellipse relates to allocation decisions, while rounded boxes are the specific GHG calculation steps. 4 5
When default published values for GHG emissions from the production of feeds are used, it is 6
important to account for their system boundary. For example, the system boundary for the default 7
values in the LEAP Animal Feed Guidelines ends at the “animal’s mouth”. When feed production 8
emissions are integrated into the calculation of emissions for the cradle-to-farm-gate, it is important to 9
ensure that double counting is avoided and that all emissions are included. 10
In practice, there is wastage of feed at various stages between harvest and storage (covered in the 11
LEAP Animal Feed Guidelines), and during feeding to animals, and this shall be accounted for. For 12
example, if there is 30 percent wastage between the amount fed to animals and the amount consumed, 13
53
the emissions from feed inputs shall be based on the amount fed. This waste feed may end up in the 1
manure management system and its contribution to subsequent methane and N2O emissions during 2
storage shall be accounted for. 3
As noted in Section 6, a large proportion of sheep and goats globally are managed in extensive 4
systems with grazing of perennial pastures or browsing on mixed forage systems, including shrubs. 5
Some important features of these feed types, in contrast to annual crops and concentrates, are 6
relatively low inputs associated with their production, lack of crop residues associated with regular 7
plant renewal, and variable feed quality throughout the year. The latter means that a single average 8
dataset will be less accurate than if a seasonal or monthly profile of plant analyses is used and linked 9
with seasonal or monthly estimates of animal feed intake. 10
The amount of feed used shall be based on the calculated intake by the animals over a one-year period. 11
Thus, for a feed that is harvested and brought to the animals (e.g. a concentrate), the annual amount of 12
dry matter feed used (plus any allowance for wastage) shall be calculated and multiplied by the 13
emissions per kg feed (i.e. kg CO2-equivalent/kg dry matter). In some extensive systems in Asia and 14
Africa, and in Australia during periods of extended drought, cuttings from trees would also represent a 15
feed that is harvested and brought to animals. In such cases, there is a need to account for any inputs 16
used in their production, as well as for the harvesting and transport of feed to the animals, to determine 17
any feed-related emissions. Where that tree has multiple uses (e.g. for fruit production for sale as well 18
as for a source of forage), the GHG emissions from the total inputs to the tree shall be allocated 19
between the two uses, or if they are a very low value waste product, no production-related emissions 20
would be allocated to them (described in the LEAP Animal Feed Guidelines). 21
Cereal straw or other plant residues may be used for bedding in housed animal systems. In such cases, 22
GHG emissions associated with the harvest and transportation steps of such products shall need to be 23
accounted for. 24
25
11.2.2 ANIMAL POPULATION AND PRODUCTIVITY 26
The calculation of animal-derived GHG emissions (e.g. enteric methane, excreta N2O) requires data on 27
total feed intake and some feed quality parameters. In most small ruminant production systems it is not 28
possible to obtain direct data on feed intake. This applies particularly to farm systems with direct 29
grazing or browsing of perennial forage plants. Thus, feed intake is commonly determined indirectly 30
using models that calculate feed requirements according to animal numbers and their productivity. 31
Most models used for calculation of feed requirements derive intake from the energy requirements for 32
animal processes of growth, reproduction, fibre production, milk production, activity 33
(i.e. grazing/walking) and maintenance (e.g. IPCC, 2006; NRC, 2007). This requires data on relevant 34
animal numbers and productivity. 35
54
To account for total GHG emissions from animal products over a one-year time period, it is necessary 1
to define the animal population associated with the production of the products (e.g. see Figure 10 for a 2
simplified sheep population example). This requires accounting for the number of breeding female and 3
male animals, replacement female and male animals, and surplus (i.e. not required for maintenance of 4
the flock) animals sold for meat. A minimum requirement for animal numbers for a stable population 5
could be the number of adult breeding animals and the number and class (i.e. age category and gender) 6
of animals sold for meat. However, it is recommended that an animal population “model” be 7
constructed from: 8
the number of adult breeding animals; 9
a herd replacement rate (from which numbers of replacement animals could be calculated); 10
fertility (i.e. lambing or kidding %, equivalent to the number of animals weaned as a 11
proportion of the number of breeding adult ewes/does, as well as lambs/kids produced from 12
growing replacements e.g. ewe hoggets); 13
death rate; 14
average age at first lambing/kidding. 15
55
FIGURE 10: SIMPLIFIED EXAMPLE OF A SHEEP POPULATION ILLUSTRATING RELATIVE NUMBERS OF BREEDING AND 1 REPLACEMENT SHEEP ON‐FARM AND SURPLUS SHEEP SOLD FOR MEAT PROCESSING 2
3
Note: Based on breeding ewe flock of 1 000, 100 percent lambing, 25 percent replacement rate, 2 percent death rate and first 4 lambing at 2 years of age. 5 6
From the base animal population data, an annual stock reconciliation needs to be derived that accounts 7
for the time of lambing/kidding and time of sale of surplus animals. Ideally, a monthly stock 8
reconciliation would be used. The benefit of having a tier-2 methodology (using calculated energy 9
requirements; see Glossary) and specific seasonal or monthly data is that the effects of improvement in 10
animal productivity on reducing the carbon footprint of products can be determined. For example, 11
achieving the final slaughter weight of lambs earlier results in a lower feed intake and the maintenance 12
feed requirement is reduced relative to the feed needed to achieve a given level of animal production. 13
The population data may need to be extended to include animals transferred between farms. For 14
example, growing “store” lambs may be sold or moved to another farm for finishing to final live-15
weight before sending off for processing. In this case, all necessary components for the production of 16
the acquired animals on the contributing farm shall be accounted for, including adult breeding stock. 17
For national or regional level analyses, this can be accounted for using average data. However, for 18
case studies it will require primary data from all source farms, and where these data are unavailable, it 19
will be necessary to use regional data for the specific animal classes on the contributing farm(s). 20
Simplifications may be necessary for minor contributors such as accounting for breeding rams or 21
bucks. These are often sourced off other farms, but can be accounted for by assuming they are derived 22
56
from within the base farm system (e.g. Figure 10). Ideally the transport component of externally 1
sourced ram/bucks would be included in the calculations. 2
Calculation of animal productivity also requires average data on male and female adult live-weight, 3
live-weight of animal classes at slaughter, fibre production and milk production (for milking sheep or 4
goats). Average birth weight is also required, but a reasonable default value for lambs is 9 percent of 5
the adult ewe live-weight or about 7 percent for goats. 6
Primary data on the animal population and productivity shall be used where possible. The minimum 7
amount of primary data to develop an animal population summary was described above, but if this was 8
unavailable then an example of sheep flock and goat herd parameters for different regions of the world 9
is given in Appendix 5. 10
11
a) Calculating energy or protein requirements of animals 12
A range of models are used internationally for estimating the energy requirements (either as net or 13
metabolizable energy) of ruminants from animal population and productivity data. Many of these have 14
similar driving functions (e.g. maintenance requirements based on body-weight0.75) with variations in 15
equation parameters according to data from specific animal metabolism studies and field validations. 16
Where country-specific models for calculating the energy requirements for sheep or goats have been 17
published, and used in the National Greenhouse Gas Inventory (NGGI) for that country, these shall be 18
used. Where alternative models (e.g. region-specific published models) are used to improve the 19
accuracy of the calculations, these should be described in detail and justified. Many groups in the 20
GHG research area use the IPCC (2006) energy requirement model; therefore, it is recommended that 21
this model be used as the main default methodology. However, the equation for energy requirements 22
for fibre growth shall be adjusted to account for the efficiency of ME requirements for fibre growth 23
using 157 MJ/kg fibre (NRC, 2007; not the current IPCC value of 24 MJ/kg fibre, which is simply 24
based on energy content of wool). It is acknowledged that use of the IPCC (2006) model requires 25
application of the sheep model for goats, and that it may give lower estimates than some other models 26
such as NRC (2007) and CSIRO (2007). The recommended order of preference is: 27
1. country-specific models used in the country NGGI; 28
2. other models that have been peer-reviewed and published that are applicable to the region and 29
country; 30
3. IPCC (2006) model; 31
4. IPCC default tier-1 values (this should be seen as a last resort). 32
57
Determination of metabolizable protein requirements by animals is required for systems where fibre is 1
an important product, so that allocation between fibre and co-products (e.g. meat) can be calculated 2
(Section 11.2.5). The above information on energy requirements also applies for protein requirements, 3
including the hierarchy of methods. However, IPCC (2006) does not include calculations for protein 4
requirements and therefore option 3 will be the NRC (2007) metabolizable protein requirement models 5
for sheep or goats. 6
7
b) Assessment of feed intake 8
In a limited number of situations, it will be possible to use measured data to define the amount of feed 9
intake on-farm to produce the animal product(s). This is only likely to apply where animals are 10
permanently housed and all feed is brought to them. However, in most cases, small ruminants obtain 11
feeds from a number of sources, including by grazing or browsing, and it is not possible to have an 12
accurate measurement of the total amount of feed consumed. In such cases, the total feed intake is 13
calculated from the total energy requirements of the animals, as outlined in Section 11.2.2. 14
Calculation of feed intake from the energy requirements of an animal that consumes a number of feed 15
types will commonly require several steps. The following describes the process using metabolizable 16
energy (ME). 17
The first step is to define the measured amount of feed intake from any supplied feed sources brought 18
into the farm from an outside source (e.g. where concentrates are provided as a supplement). This 19
needs to account for the total amount of the particular feed(s) provided and adjust for the level of feed 20
consumption and wastage (i.e. using a utilization percentage). Some examples of losses by wastage are 21
5-10 percent when feed is provided to animals in specialized feeding facilities through to 20-22
40 percent when fed by spreading out on ground or pasture (DairyNZ, 2012). The first calculation step 23
will involve subtracting the amount of ME consumed from the supplied feed(s) (based on the amount 24
of feed DM intake and its specific energy concentration in MJ ME/kg DM) from the total energy 25
requirements to determine ME intake from other feed source(s): 26
27
MEintakeother = Total ME requirement – (DMintake x MJ ME/kg DM)feed1 – (DMintake x MJ 28
ME/kg DM)feed2 29
30
The difference (MEintakeother) will be the amount of energy consumed from other feed sources, for 31
example, from grazing pasture or browsing a mix of shrubs and forages. If there is one source 32
(e.g. pasture), then the amount of DM intake from that source can be calculated (based on its specific 33
energy concentration in MJ ME/kg DM) from: 34
58
DM intakeother = MEintakeother / (MJ ME/kg DM)other 1
2
If there is more than one other feed source, it will be necessary to determine the DM intake for each 3
source from an estimate of the proportion of each feed type consumed and their specific energy 4
concentrations in MJ ME/kg DM. 5
For each feed source utilized by sheep or goats, there is a need to have an accurate average estimate of 6
the concentrations of dry matter, metabolizable energy (MJ ME/kg DM), digestibility (kg digestible 7
DM/kg total DM) and nitrogen (g N/kg DM). While these will be necessarily averaged values, the 8
most accurate data available for the specific system or regional system should be used. The latter two 9
parameters are used in the calculations of emissions from excreta for methane and N2O, respectively. 10
These feed compositional parameters can be obtained from feed measurements from the farm 11
system(s) studied, average published data relevant to the agro-ecological zone being studied, or 12
published national or global data for the relevant feeds. For forage or browse species that show 13
marked seasonal variation in quality, seasonal data (or monthly data if available) should be used where 14
possible. Default annual average data for a wide range of different feed sources are given in NRC 15
(2007). 16
17
c) Animal enteric methane emissions 18
According to IPCC (2006), an average of 6.5 percent (±1%) of gross energy intake is lost as enteric 19
methane from the rumen of mature sheep and 4.5 percent from lambs (< 1 year old). Goats are 20
assumed to have the same methane loss factors. Data for cattle generally indicate that this loss factor is 21
higher for lower digestibility feeds, but there are limited data for development of scaling factors for 22
small ruminants and therefore the single emission factors shall be used. 23
Determination of the amount of methane emitted from each animal class requires multiplication of the 24
total net energy (NE) or ME intake by each animal class (as described in Section 11.2.2) by methane 25
emission factors. However, a first step is the conversion of total NE (or ME) intake to gross energy 26
intake, using data on feed percentage digestibility (see Appendix 6; IPCC, 2006). The annual quantity 27
of methane emitted for each animal class is then calculated using the following equations: 28
29
kg CH4/mature animal/year = gross energy intake (MJ/year) x 0.065 / 55.65 30
kg CH4/animal(<1 year-old)/year = gross energy intake (MJ/year) x 0.045 / 55.65 31
59
The values of 0.065 and 0.045 refer to the 6.5 percent and 4.5 percent loss factors for methane of gross 1
energy intake, while 55.65 is the energy content of methane in MJ/kg. 2
Thus, annual enteric methane emissions per animal per year are calculated using the above equation 3
and the gross energy intake for one year for each animal class and integrating them up across the 4
number of animals. This represents a default international emission approach based on tier-2 5
methodology. Where country-specific emission factors have been peer reviewed, published and 6
integrated into the national GHG Inventory, then these shall be used instead .If a user of these 7
guidelines is unable to access sufficient basic data to apply the above approach, then a tier-1 emission 8
factor could be used based on the IPCC (2006) default values of 8.5 kg and 5 kg CH4/animal/year for 9
sheep in developed countries, and sheep in developing countries and goats, respectively (based on 10
live-weights of 65 kg, 45 kg and 40 kg, respectively). However, the use of tier-1 factors mean that the 11
user has no ability to account for carbon footprint reductions associated with improvements in animal 12
productivity. 13
14
11.2.3 MANURE PRODUCTION AND MANAGEMENT 15
a) Methane emissions from animal excreta and manure 16
Methane emissions from animal excreta and manure are estimated by first calculating the amount of 17
volatile solids produced. This represents the amount of feed consumed corrected for the component 18
digested by animals and the non-volatile ash component that remains. The equations for calculating 19
volatile solids in IPCC (2006; Equation 10.24) can be simplified to: 20
21
kg volatile solids = [kg DM intake by animals x (1.04 – DMD)] x 0.92 22
23
where DMD is the dry matter digestibility expressed as a fraction. For example, the % DMD for 24
perennial pastures in New Zealand varies throughout the year, from about 74 percent in summer to 25
84 percent in winter (Pickering et al., 2011). The 0.92 factor in the above equation is based on a 26
default of 8 percent ash content of manure (i.e. using 1 – [%ash/100]), which should be modified if 27
measured or if country-specific values differ from this default. 28
The methane emission factor calculations for the volatile solids vary according to the manure 29
management system and climate (IPCC, 2006). If the tier-2 approach cannot be used, generic tier-1 30
emission factors are given by IPCC (2006; Table 10.15) for sheep and goats in developed or 31
developing countries, and different temperature regimes. Where country-specific emission factors 32
have been peer reviewed, published and integrated into the national GHG Inventory, then these shall 33
be used instead 34
60
b) Nitrous oxide emissions from animal excreta and manure 1
Nitrous oxide emissions occur by direct emissions from excreta, indirectly from ammonia released 2
from excreta into the atmosphere and deposited back onto soil, and from nitrate leached to waterways. 3
A tier-2 approach shall be used whereby the amount of N excreted by animals is calculated using the 4
animal production and feed intake model outlined in Sections 11.2.2a–11.2.2.b. The amount of DM 5
intake is multiplied by the average N concentration of the diet (weighted according to the relative 6
proportions of different feed types “t” in the diet) to get the amount of N consumed: 7
8
kg N consumed = Σ (kg DM intaket x %N in feedt /100) 9
10
N output in product(s) (i.e. meat, fibre, milk) is then subtracted from the N consumed to calculate the 11
amount of N excreted: 12
13
kg N excreted = kg N consumed – kg N in products 14
15
Data on the average N concentration of a wide range of different feed sources is given in the LEAP 16
Animal Feed Guidelines and NRC (2007), but this shall be over-ridden by measured values 17
(i.e. primary data) or country-specific peer-reviewed published values, if available. The N output in 18
products is calculated from the amount of product multiplied by the protein concentration of the 19
product and divided by 6.25 to convert protein to N: 20
21
kg N in products = [kg product x (% protein in product / 100) / 6.25] 22
23
The values for protein concentration of products should be based on measured values or country-24
specific peer-reviewed published values, where possible. Typical default values for the protein 25
concentration of meat (live-weight gain basis), clean wool (dry weight basis; scoured wool typically 26
has about 16 percent water) and milk from sheep are 21 percent, 100 percent and 5.8 percent, 27
respectively (e.g. Pulina et al., 2005). Corresponding typical values for goats for meat, clean fibre and 28
milk are 21 percent, 100 percent and 3.0 percent, respectively. 29
Direct N2O emissions from excreta deposited on soil during grazing or browsing are calculated by 30
multiplying the annual amount of N excreted by the IPCC (2006) emission factor of 0.01 kg N2O-N/kg 31
N excreted (see Figure 11 for a summary of calculation components). Where country-specific 32
61
emission factors have been published and integrated into the national GHG Inventory, then these shall 1
be used instead. When excreta is collected and processed via a manure management system, the 2
storage-related emissions are to be included in this analysis. Where the stored manure is transported 3
away and applied to land growing a crop or pasture, the emissions associated with transport and 4
application (after adjustment for N lost by volatilization) are found in the LEAP Animal Feed 5
Guidelines and not within the current LCA Guidelines. 6
For calculation of N2O emissions from manure during storage, the relevant IPCC (2006) emission 7
factors shall be used. For example, direct N2O emission factors in kg N2O-N/kg N from storage vary 8
from nil for uncovered anaerobic lagoons, 0.005 to 0.01 from aerobic ponds (being less with forced 9
aeration), 0.02 from drylot to 0.1 for composting with regular turning and aeration (IPCC, 2006; 10
Table 10.21). 11
Indirect N2O emissions from ammonia emissions during storage first require an estimate of the amount 12
of ammonia emitted. This can be calculated using country-specific factors that have been published 13
and integrated into the national GHG Inventory or the IPCC (2006) default ammonia loss factors 14
(FRACGASM) for excreta-N deposited in deep-bedding or solid storage animal-housing manure 15
management systems of 25 percent and 12 percent, respectively. Ammonia-N loss is then multiplied 16
by the IPCC (2006) emission factor (EF4) of 0.01 kg N2O-N/kg N excreted. 17
Indirect N2O emissions from ammonia loss and N leaching from excreta deposited directly to land 18
during grazing shall be calculated as shown in Figure 11. Country-specific factors that have been 19
published and integrated into the national GHG Inventory shall be used and, if not available, the IPCC 20
(2006) default factors shall be used. Calculations first require an estimate of the amounts of ammonia 21
loss and N leaching from excreta deposited on land. The default IPCC (2006) loss factor for 22
FRACGASM is 20 percent of N excreted, and for FRACLEACH is 30 percent (for soils with net drainage, 23
otherwise 0 percent) of N excreted. These are then multiplied by the corresponding IPCC (2006) 24
emission factors [EF4 and EF5] of 0.01 kg N2O-N/kg N lost as ammonia and 0.0075 kg N2O-N/kg N 25
leached, respectively. 26
The total N2O emissions from excreta and manure are calculated by summing the direct and indirect 27
N2O emissions, after adjustment for the N2O/ N2O-N ratio of 44/28. 28
62
FIGURE 11: SUMMARY OF APPROACH FOR CALCULATING N2O EMISSIONS FROM ANIMAL EXCRETA AND THE 1 ANIMAL WASTE MANAGEMENT SYSTEM (AWMS) USING IPCC (2006) ACTIVITY FACTORS (FRAC REFERS TO 2 FRACTION OF N SOURCE CONTRIBUTING) AND EMISSION FACTORS (EF IN KG N2O‐N/KG N) 3
4
Note: GASM = gaseous loss as ammonia. FRACgasm and EF1 vary with type of AWMS. For manure, only manure storage 5 losses are included in these guidelines (losses from land application are covered in the LEAP Animal Feed Guidelines). 6 7
11.2.4 EMISSIONS FROM OTHER FARM‐RELATED INPUTS 8
The other main inputs on-farm that contribute to GHG emissions are largely associated with the use of 9
fuels and electricity. Additional farm-related inputs that need to be accounted for include consumables 10
used on-farm. Nutrients administered directly to animals and milk powder used for rearing lambs or 11
kids are covered in the LEAP Animal Feed Guidelines. 12
The total use of fuel (diesel, petrol) and lubricants (oil) associated with all on-farm operations shall be 13
estimated. Estimations shall be based on actual use and shall include fuel used by contractors involved 14
in on-farm operations. Where actual fuel-use data are unavailable, these should be calculated from the 15
operating time (hours) for each activity involved in fuel use and the fuel consumption per hour (this 16
latter parameter can be derived from published data or from appropriate databases, such as Ecoinvent). 17
Note that any operations associated with the production, storage and transportation of animal feeds are 18
not included here, but are covered in the LEAP Animal Feed Guidelines. Figure 8 indicates some of 19
the main non-feed processes associated with the use of fuels, such as water transport, use of vehicles 20
for animal movement, removal of wasted feed associated with provision of feeds to animals on-farm 21
and other farm-specific activities (e.g. visit by veterinarians). 22
63
The total amount use of a particular fuel type is then multiplied by the relevant country-specific GHG 1
emission factor (which accounts for production and use of fuel) to determine fuel-related GHG 2
emissions. The process for calculating fuel-related GHG emissions also applies to electricity. Thus, all 3
electricity use associated with farm activities (excluding feed production and storage where they are 4
included within the emission factor for feeds) shall be estimated. This includes electricity for water 5
reticulation, animal housing, milk collection (for milking goats and sheep) and shearing of fibre from 6
animals (Figure 8). Country-specific emission factors for electricity production and use shall be 7
applied according to the electricity source. This would typically be the national or regional average 8
and would account for the electricity grid mix of renewable and non-renewable energy sources. 9
In some extensive production systems, nutrients required to avoid deficiency by animals 10
(e.g. phosphorus, magnesium, sodium) may be delivered directly to animals rather than being applied 11
to land for uptake by plants used as animal feeds. In such cases, this may represent a significant 12
contribution to total GHG emissions and shall be accounted for, as described in the LEAP Animal 13
Feed Guidelines. 14
Where there is a significant use of consumables in farm operations, the GHG emissions associated 15
with their production and use should be accounted for. This would generally be estimated from 16
published data or from appropriate databases (e.g. Ecoinvent). However, in practice these will often 17
constitute a very minor contribution and relevant data on them may be difficult to access. See 18
Section 8.4.3 on cut-off criteria for exclusion of minor contributors. 19
20
11.2.5 MULTI‐FUNCTIONAL PROCESSES AND ALLOCATION OF GHG EMISSIONS BETWEEN CO‐PRODUCTS 21
a) Accounting for different animal species and non‐feed activities within a farm 22
Many farms present a mixture of animal species (e.g. sheep, goats, cattle, deer) which are often farmed 23
together. Where possible, it is recommended to separate activities of the farm system for the different 24
animal species where specific uses can be defined (e.g. use of summer forage crops to finish lambs 25
only, or use of nitrogen fertilizer specifically for pasture growth to feed beef cattle). For the remainder 26
of the GHG emissions for the cradle-to-farm-gate stage, where there is common grazing or feeding of 27
the same feed source, the actual amount of feed consumed by the sheep or goats under study shall be 28
calculated as outlined in Section 11.2.2. Emissions associated with other non-feed shared activities 29
(e.g. fuel used for animal movement, drain cleaning, hedge cutting, fencing maintenance, etc.) shall be 30
allocated between animal species using a biophysical allocation approach. Preferably, this should be 31
based on calculation of the total feed intake for each of the different animal species and allocation 32
based on the relative feed intake between species (see Box 1). 33
64
BOX 1: EXAMPLE OF CALCULATION OF MULTI‐FUNCTIONAL PROCESSES AND ALLOCATION IN A FRENCH MIXED 1 SHEEP AND CATTLE FARM 2
3
The figure above describes the farm system (based on Benoit and Laignel, 2011). The area identified as being 4 used for cash crops is excluded in the calculation of GHG emissions from animals on-farm. The main fodder 5 area is pasture (in white), which is commonly grazed and used for silage or hay production for both sheep and 6 beef cattle. The table below describes a process used in France to apportion GHG emissions between cash crops 7 and animal species for the Case Study farm. 8
9
Source/stage of co‐products Recommended method Basis
1st: Split between cash crops and animal production (including crops for animals and forages)
Fuel Total fuel use only French empirical references (Litres/ha and
litres/LU) used to build specific allocation
keys
Electricity Total electricity only, except for
specific usages (irrigation)
French empirical references (kWh/LU)
used to build specific allocation keys
Manure fertilizers Amounts known for each crop and
forages
Split between cash crop and feeds for
animals, i.e. system separation Manure application
2nd: Then split between the different types of animal production
Forages (production and
conservation for silage or hay
[e.g. including plastics])
General data on forages only Biophysical allocation (based on relative
feed intake) for forages (pasture, silage,
hay) used by both animal species
Cereal crops and maize silage for
animals
Quantities distributed to each animal
species are known
System separation
Feed inputs (concentrates, vitamins,
minerals, milk powder)
Quantities (or amount in €) distributed
to each animal species are known
System separation
Breeding operations
(e.g. reproduction, veterinary,
drenches)
Can be assessed through economic
value, but are known for each animal
type
System separation
10
11
65
Total fuel use is known, but this is used for production of cash crops, feeds for animals and general farm 1 activities relating to animals (e.g. provision of feed, removal of feed waste, manure management, vehicles for 2 animal movements). French researchers allocate fuel-related emissions between cash crops and each animal 3 type using empirical functions derived from regional survey data and related to hectares of crop or livestock 4 units (where a livestock unit or LU equates to one cattle eating 4 750 kg DM; see table below). 5 6
7 An alternative approach for fuel is to use records of all specific farm operations relating to each crop 8 (e.g. hectares ploughed, rotary-tilled, sown, harvested), then use country-specific or published values for 9 typical fuel use per hectare (e.g. Witney 1988) and integrate these for each system, thereby allowing a system 10 separation approach. In this case, biophysical allocation would then be applied for the remaining fuel used for 11 pasture-related activities and non-feed animal activities (e.g. manure management, animal movements) to 12 allocate it between sheep and cattle (see below). 13 14 A similar approach is used for electricity use in France based on a database of average use for sheep, cattle or 15 cropping [0.4 GJ/LU or 0.4 GJ/ha]. Alternatively, a biophysical allocation ratio could be applied to allocate 16 between animal type (see below). 17 18 System separation can be used for the main crops, other feed sources and animal breeding operations (see table 19 above). However, the sheep and cattle jointly graze pasture on the farm and are jointly fed pasture silage and 20 hay. Therefore, some method is required for apportioning the related inputs and emissions between sheep and 21 cattle. The simplest biophysical allocation method is to use the total energy requirements (or DM intake) for 22 sheep and cattle. In this case, the allocation factor (A) was calculated using: 23 24
A (%) = 100 x Sheep total DM intake/ (Sheep total DM intake + Cattle total DM intake) 25 26 In this farm, A = 100x347/(347+190) = 65% (where 347 and 190 are t DM intake calculated for sheep and 27 cattle respectively). Thus, 65 percent of farm-related GHG emissions (or fossil energy demand) that could not 28 be separately estimated or derived via system separation would be attributed to sheep. 29
30 b) Allocation between meat, milk and fibre production 31
Small ruminants produce meat, milk and fibre. However, in most production systems the focus is on 32
one main product, or one product may provide the largest proportion of economic returns for the 33
producer. For dairy cows where the main product is milk and where there is a meat co-product, a 34
biophysical allocation approach is most widely used (e.g. Thoma et al., 2013a), and is recommended 35
in the IDF (2010a) methodology. The same approach shall be used for sheep or goats where the main 36
product is milk. Using a tier-2 approach, the energy requirements for milk and meat production are 37
calculated according to an internationally acceptable methodology (Section 11.2.2.a; since energy is 38
the main determining requirement for these products). The allocation ratio for milk, relative to milk 39
Sheep production Cattle production Cash crops Total
References
in buildings
on areas
1 GJ/LU
+ 0.9 GJ/ha of fodder area +0.4GJ/ha crop
1.8 GJ/LU
+ 1.4 GJ/ha of fodder area +0.4GJ/ha crop
4.3 GJ/ha
Theoretical consumption
1*73LU +0.9*56
= 123.4
1.8*40LU +1.4*36
= 122.4
4.3*34ha
=146.2
392
Allocation % 31.5 31.2 37.3 100
66
plus meat (plus fibre if it is also a minor co-product), is then calculated from the ratio of the energy 1
requirement for milk production to the energy requirement for milk, meat (i.e. animal growth 2
component) and fibre (if relevant) production (see Box 2): 3
4
Allocation % to milk = 100 x [energy req. for milk/(energy req. for milk + energy req. for meat + 5
energy req. for fibre)] 6
7
BOX 2: EXAMPLE OF CALCULATION OF BIOPHYSICAL ALLOCATION BETWEEN MILK AND MEAT FOR GOATS IN KENYA 8 9 This example refers to a dual-purpose goat system based on 100 breeding does producing 8 775 kg milk sold 10 (and the same amount being fed to kids) and 3 440 kg live-weight sold for meat (from surplus kids and cull does 11 and bucks). It was based on data from Bett et al. (2009). 12
A goat population structure was prepared and production data was used to calculate energy requirements using 13 the IPCC (2006) model. Results for the goat population showing the relative breakdown between functions are 14 given in the figure below. 15
16 The ratio (R) of energy requirements for sold-milk production (half of the total was sold, with the rest fed to kids) 17 relative to meat production (based on the requirements for growth) was calculated using: 18
R = energy req. sold-milk/(energy req. sold-milk + energy req. growth) = 40.3/(40.3+17.4) = 0.70 19
Thus, the allocation percentage for milk and meat was 70 percent and 30 percent, respectively. 20
For sensitivity analysis of the effect of allocation method, the corresponding allocation percentage for milk using 21 economic allocation was 45 percent and using protein-mass was 50 percent. 22
23
This equation is rearranged to calculate the corresponding allocation % to meat and fibre. Where milk 24
or meat is the main product from the production system, biophysical allocation based on energy 25
requirements shall be used. However, where fibre is an important product, biophysical allocation 26
based on protein requirements shall be used. The latter recommendation is based on the fact that fibre 27
production is largely determined by protein requirements and not energy requirements. Additionally, 28
most energy-based models are relatively weak in accounting for fibre production and simply include 29
fibre production within the energy requirements for maintenance, or use a crude estimate of energy 30
67
content of fibre (e.g. IPCC, 2006; in Section 11.2.2.a, it is recommended that the energy requirement 1
for fibre is changed to 157 MJ/kg fibre). 2
Thus, where fibre is an important co-product the allocation ratio for fibre, relative to fibre plus meat 3
(plus milk if it is also a minor co-product), is then calculated from the ratio of the metabolizable 4
protein requirement for fibre production to the metabolizable protein requirement for fibre, meat 5
(i.e. animal growth component) and milk (if relevant) production using: 6
7
Allocation % to fibre = 100 x [protein req. for fibre/(protein req. for fibre + protein req. for meat + 8
protein req. for milk)] 9
10
This equation is rearranged to calculate the corresponding allocation percentage to meat and milk co-products. 11
Box3 provides an illustration of application of this approach for a New Zealand farm system. 12
13
BOX 3: EXAMPLE OF CALCULATION OF BIOPHYSICAL ALLOCATION BETWEEN FIBRE AND MEAT FOR SHEEP IN NEW 14 ZEALAND 15
In North Island, New Zealand, Romney-cross sheep graze on hill country with cattle. On average, for this farm 16 class the sheep are stocked at 5.6 sheep/ha/year and produce 207 kg live-weight sold/ha/year and 30.8 kg 17 wool/ha/year (Beef + Lamb New Zealand survey data). It is based on data from Ledgard et al. (2009, 2011). 18
Farm survey data were used to define a sheep population structure and sheep production. The data were 19 incorporated into a protein requirement model (based on the Australian Feeding Standards model; CSIRO 2007) 20 to determine protein requirements. 21
Of the total metabolizable protein requirements, 17 percent was required for wool production and 29 percent for 22 sheep growth, pregnancy and lamb production. The remaining protein requirement was for flock maintenance. 23
The ratio (R) of protein requirements for wool production relative to meat production (based on the total 24 requirements for growth) was calculated using biophysical allocation as: 25
R = protein req. wool / (protein req. wool + protein req. growth) = 17/(17+29) = 0.37 26
Thus, the allocation percentages for wool and meat were 37 percent and 63 percent, respectively. 27
Sensitivity analysis was carried out to examine the effect of other methods of allocation. This resulted in 28 calculated values for the allocation percentage for wool using biophysical allocation based on energy 29 requirements (and using 157MJ/kg wool) or economic allocation of 16 and 19 percent, respectively. 30
31
In practice, there are relatively few small ruminant production systems where fibre is the main product 32
and returns the highest proportion of revenue. These include specialist ultra-fine wool sheep in 33
Australia and some goat systems for specialist cashmere and mohair production. It is acknowledged 34
that the use of a biophysical allocation approach based on protein requirements for fibre, meat and 35
68
milk is still in a state of development (with no scientific publications on the topic). On this basis, use 1
of a protein requirement model may not always be possible. Therefore, a less-preferred alternative 2
option is use of energy requirements only. The recommended hierarchy for calculating biophysical 3
allocation is: 4
1. where milk is the dominant product this shall be based on energy requirements, but where 5
fibre is a dominant or important co-product it shall be based on protein requirements; 6
2. use energy requirements for all product evaluations. 7
If the second option is used, it shall be accompanied by an explanation and justification of the method 8
used for the assessment. 9
In conformance with ISO/TS 14067 (2013), where choice of allocation can have a significant effect on 10
results, it is recommended to use more than one method to illustrate the effects of choice of allocation 11
methodology. For example, system expansion, protein mass or economic allocation should be used for 12
comparison, with the latter based on the relative gross economic value of the products received 13
(e.g. using regional/national data) over a period of at least three years (to reduce potential effects of 14
price fluctuations over time). 15
16
11.3 Transportation 17
This section refers to transportation stages covering transport of animals, fibre or milk from the site of 18
production to the site of primary processing, and any internal transport within the primary processing 19
site(s) to the output loading dock (see Section 11.6). It also includes transportation of inputs such as 20
water within the farm, and animals between different farms, contributing to their production prior to 21
going for processing. 22
Fuel consumption from transport can be estimated using: (i) the fuel cost method, (ii) the fuel 23
consumption method, or (iii) the tonne-kilometre method. When using the fuel cost method (fuel use 24
estimated from cost accounts and price) or the fuel consumption method (reported fuel purchased), the 25
“utilization ratio” of materials transported shall be taken into account. Transport distances may be 26
estimated from routes and mapping tools or obtained from navigation software. Regarding return 27
empty transport, allocation of empty transport kilometres shall be done on the basis of the average 28
load factor of the transport, representative for the transport under study. If no supporting information 29
is collected, it should be assumed that 100 percent extra transport is needed for empty return. 30
Allocation of transport emissions to transported products shall be done on the basis of the mass share, 31
unless the density of the transported product is significantly lower than average such that the volume 32
restricts the maximum load. In the latter case, it shall be done on a volume basis. When cold chain is 33
used, life cycle emissions from cold and frozen storage shall be collected – including refrigerant loss. 34
69
Where live exports of animals occur (e.g. by shipping to middle-Eastern countries), it is necessary to 1
account for all related transport emissions and loss of animals during transportation. The use of fuels 2
and GHG emission factors associated with the type of transportation shall be calculated according to 3
the size of transportation vehicle and the typical fuel consumption rate. Where refrigerated 4
transportation is used, the typical rate of loss of refrigerant and associated GHG emission factor shall 5
be included. 6
7
11.4 Inclusion and treatment of landusechange 8
Land-use-change (LUC) relates to the feed production stage and is therefore covered in the LEAP 9
Animal Feed Guidelines. These guidelines describe two calculation methods, including a global 10
averaging method if specific land use details are unknown and where LUC effects are spread across all 11
land use. Calculations using the latter method shall exclude long-term perennial forages such as 12
perennial pastures, rangeland and browse systems (i.e. global average LUC GHG is zero). long-term 13
perennial forage systems can be significant in some small ruminant systems. GHG emissions 14
associated with LUC should be accounted separately and reported. PAS 2050 (BSI, 2011) provides 15
additional guidance. 16
17
11.5 Biogenic and soil carbon sequestration 18
Biogenic and soil carbon sequestration can be important for some small ruminant systems. However, 19
since this relates only to the feed production stage, the specific methods are covered in the LEAP 20
Animal Feed Guidelines. As these guidelines note, biogenic and soil carbon sequestration shall be 21
included in the final GHG emissions value. Where no data relating to soil carbon sequestration are 22
available, the LEAP Animal Feed guidelines provide default values for temperate climate. The last 23
option is to assume zero change in soil carbon. 24
25
11.6 Primary processing stage 26
The three primary products of milk, fibre and meat are covered in these guidelines. For all products, 27
there are a number of generic processes that contribute to GHG emissions. These are summarized in 28
Figure 12 and include transportation (of products within or between primary processing plants), 29
processing, water use, cold/frozen storage, and wastes and wastewater treatment. Each component 30
requires raw materials associated with production of energy carriers, refrigerants, consumables, 31
cleaning chemicals and packaging. The following sections discuss the specific products and the 32
assessment of GHG emissions with their primary processing. 33
34
70
FIGURE 12: PROCESSES THAT CONTRIBUTE TO GHG EMISSIONS AND FOSSIL ENERGY DEMAND WITHIN THE 1 SYSTEM BOUNDARY OF THE CRADLE‐TO‐PRIMARY‐PROCESSING‐GATE 2
3
Note: Related cradle-to-farm-gate processes are also given (a further breakdown of these is given in Figure 8). The box with a 4 blue background refers to inputs, processes and emissions covered by the LEAP Animal Feed Guidelines and not part of the 5 current guidelines. 6
71
11.6.1 MILK PROCESSING 1
The milk collected from goats or sheep may be used to produce one or more of the following products: 2
fresh milk, yoghurt, cheese, cream/butter, whey and milk powder. A very diverse range of products are 3
produced during processing and a wide range of technologies are used for their production, from 4
cottage industry to large multi-process facilities. IDF (2010a) and DairyCo (2010) provide an outline 5
of processes and methods for LCA-based carbon footprinting of dairy cow milk, but there are no 6
corresponding reports or PCRs for goat or sheep milk processing. 7
The main processes that need to be accounted for are processing of milk, production and use of 8
packaging, refrigeration, water use and wastewater processing, and within-plant transportation 9
(Figure 12). The milk-processing stage includes the use of resources including energy, water and 10
consumables (e.g. detergents, cleaning chemicals). 11
12
a) Data collection and handling of co‐products 13
Representative data need to be collected from the milk-processing plant(s) for the defined one-year 14
period on the amount of milk (and its fat and protein content) entering the plant and the amount and fat 15
and protein content of the different products produced. A material flow diagram of milk input and 16
output products should be produced to account for a minimum of 99 percent of the fat and protein. 17
Representative data also need to be collected on the resources used for processing. Ideally this would 18
be collected for each unit process so that it can be allocated according to the products produced. 19
However, these are rarely available. In some cases, data may be available that can be attributed to 20
production of one specific product. In such cases, these process data should first be separately 21
assigned to the specific product before applying an allocation methodology to the remaining data. In 22
most cases it is only possible to obtain data for a whole processing plant, and in such cases a method 23
for allocation of resource use and emissions between the products is required. The IDF (2010a) has 24
defined a physico-chemical method for allocation between dairy cow milk co-products, but this is 25
untested for goat or sheep milk. Thus, a simpler approach is recommended for used based on the 26
relative fat + protein content of the co-products produced (Thoma et al., 2013a). 27
Packaging is generally a relatively small contributor to total GHG emissions (<1%) and, where this is 28
the case, secondary data are often used where no specific on-site production data are available. When 29
packaging is manufactured off-site, the calculated GHG emissions should include the production of 30
the packaging as well as the raw materials. Where glass bottles are used for liquid milk, the rate of re-31
use should be accounted for in the calculations. The guidelines produced by DairyCo (2010) provide 32
useful information related to the range of packaging materials and factors to include in the calculation 33
of GHG emissions associated with their production and use. Similarly, many other consumables and 34
cleaning chemicals are used in the processing of dairy products, and secondary data sources from 35
72
databases such as Ecoinvent will generally be used for their production and use. This also applies to 1
refrigerants, although use of primary activity data on the type and amount of refrigerants used is 2
desirable. 3
4
b) Calculating GHG emissions from milk processing 5
Activity data are required on the amounts of the various resources used. Energy use must account for 6
the type of energy. Similarly, the type of packaging materials and refrigerant(s) used should be 7
identified. The activity data are then combined with relevant GHG emission factors to calculate the 8
total emissions. For refrigerants, Forster et al. (2007, Table 2.14) provide a list of GWP (100-year) 9
factors for a wide range of refrigerants, which should be updated to coincide with future revisions by 10
IPCC. 11
Data are required on the quantities of wastewater produced, its composition and the method of 12
processing (e.g. anaerobic ponds, aerobic ponds, land application). The method of processing will 13
determine the GHGs produced (e.g. methane from anaerobic treatment systems). Emission factors for 14
methane and N2O for the different wastewater processing systems are given in IPCC (2006). 15
Total GHG emissions are calculated from the sum of all contributing sources and converted to CO2-16
equivalents according to the latest GWP factors from IPCC. The calculation of total GHG emissions 17
must include adjustment for allocation between the various co-products, as outlined in Section 9.3. 18
19
11.6.2 FIBRE PROCESSING 20
The fibre collected from goats (i.e. cashmere, mohair) or sheep (i.e. wool) may be used for a wide 21
range of purposes, including clothing, carpet-making and housing insulation. Again, there is a wide 22
diversity in the processes used to produce these products ranging from cottage industry to large-scale 23
commercial processing. An outline of processes and methods for LCA-based carbon footprinting are 24
covered in draft PCRs on textile yarn and thread of natural fibres (EPD, 2012), and woven fabrics of 25
wool or animal hair (EPD, 2011) that are relevant for cashmere and wool processing. 26
The present guidelines refer only to the primary processing stage of scouring or cleaning of the fibre. 27
The cleaned fibre may then go on to other stages, such as yarn-making for textiles (with secondary 28
processing or for direct use by a consumer), carpet-making, or low-value uses such as insulation or 29
absorbents. 30
Fibre collected from small ruminants often contains grease, suint, soil and plant material (which for 31
wool can amount to 20-40 percent of the weight of the raw greasy wool), and this is commonly 32
removed by scouring to produce clean fibre. During the scouring process lanolin may be extracted, as 33
it represents a useful co-product. The main processes that need to be accounted for in fibre scouring 34
73
are the use of cleaning chemicals (e.g. detergents, bleaching agents and acids), water, within-plant 1
transportation and wastewater processing (Figure 12). The primary processing stage includes the use 2
of resources such as energy, water and consumables. 3
4
a) Data collection and handling of co‐products 5
Representative data need to be collected from the fibre-scouring plant(s) for the defined one-year 6
period on the amount of greasy fibre entering the plant and the amount of clean fibre, lanolin and 7
residue (vegetable matter and dirt, which usually goes to waste) generated. A material flow diagram of 8
input and output products should be produced to account for a minimum of 99 percent of the mass. 9
Data on all resources used shall be collected, with important resources being electricity, water and 10
cleaning chemicals. The main energy resource used is usually electricity. Primary data on electricity 11
use shall, therefore, be collected. Where possible, the data should be collected to allow allocation 12
according to the products produced (e.g. electricity use for final purification of the lanolin). Where 13
data are only available for a whole processing plant, a method for allocation of resource use and 14
emissions between the products is required. In most cases, the clean fibre and lanolin can be 15
considered as the main products, with the residue as a valueless waste. However, in some cases the 16
residue may be further processed to a conditioner or fertilizer, and in such cases should be treated as a 17
co-product. 18
A relatively large volume of wastewater can be generated during the scouring process and data shall 19
be collected on the volume and composition (e.g. Chemical Oxygen Demand and nitrogen 20
concentration) of the wastewater and the method of wastewater processing. 21
To account for the lanolin co-product a system expansion approach could be used whereby any 22
product displaced by the lanolin could be determined. However, there are no published studies to 23
guide such an analysis. For consistency with Figure 8 and the other LEAP Guidelines, economic 24
allocation between co-products is recommended. In practice, the recovery of lanolin from greasy wool 25
amounts to about 2-7 percent by weight (higher for finer wool), therefore, most of the resource use and 26
GHG emissions will be allocated to the wool. Goat fibres generally have a lower grease content than 27
sheep wool, therefore, lanolin becomes a minor co-product and possibly a residual if <1% overall 28
value (see Figure 7). 29
30
b) Calculating GHG emissions from fibre processing 31
The total GHG emissions are calculated from the sum of the emissions associated with resource use 32
and wastewater processing. Calculation of GHG emissions associated with electricity use shall 33
account for total embodied emissions, recognizing the relative energy sources used for electricity 34
74
production in the country where the primary processing occurs. Emissions from use of other resources 1
shall also be determined using activity data and relevant GHG emission factors to calculate the total 2
emissions. 3
Data on wastewater quantity and composition are used, in conjunction with GHG emission factors 4
according to the method of wastewater processing (IPCC, 2006), to calculate GHG emissions from 5
wastewater processing. 6
Total GHG emissions shall be allocated between the various co-products, as outlined in Section 9.3. 7
8
11.6.3 MEAT PROCESSING 9
Primary processing of sheep or goats for meat production can occur in facilities ranging from 10
backyards to large-scale commercial processing abattoirs. This can result in a wide range of co-11
products, including hides (for leather), tallow (e.g. for soap, biofuel), pet food, blood (e.g. for 12
pharmaceutical products), fibre and renderable material (e.g. for fertilizer). 13
A PCR (Boeri, 2013) has been produced for generic meat processing, where the core functional unit is 14
1 kg of meat (fresh, chilled or frozen), and includes details on accounting for cold and frozen storage. 15
It also covers upstream and downstream processes, including the use phase (meat cooking). 16
The present guidelines refer to primary processing for fresh, chilled or frozen meat, and do not account 17
for secondary processing (e.g. further processing of meat into ready-to-cook dishes) or subsequent 18
retail, use and waste stages, which would be included in a full “cradle-to-grave” LCA. The main 19
processes that need to be accounted for are: animal deconstruction into many component parts, 20
production and use of packaging, refrigeration, water use and wastewater processing, and within-plant 21
transportation (Figure 12). The meat-processing stage involves the use of resources including energy, 22
water, refrigerants and consumables (e.g. cleaning chemicals, packaging and disposable apparel). 23
24
a) Data collection and handling of co‐products 25
Representative data need to be collected from the meat-processing plant(s) for a recent representative 26
one-year period on the amount of sheep or goat live-weight entering the plant and the amount of 27
different products produced. A material flow diagram of input and output products should be produced 28
to account for a minimum of 99 percent of the mass. While primary data shall be used for meat, they 29
may not be available for the relative quantity of all co-products (e.g. blood, gut contents), and 30
therefore secondary data would be required, or could be aggregated across several minor co-products. 31
An indication of the approximate relative weight of products for a lamb is: meat 52 percent, hide 32
6 percent, wool 4 percent, blood 5 percent, tallow 3 percent and renderable material 30 percent (see 33
also Box 4). 34
75
Data are required on the use of the various resources. Energy use is a major contributor to total GHG 1
emissions for the processing stage. Therefore, it is important to obtain primary data on the various 2
sources of energy use. Similarly, water use can be relatively large and wastewater processing can 3
represent a large component of the processing GHG emissions. Thus, data shall be collected on the 4
volume and composition (e.g. Chemical Oxygen Demand and nitrogen concentration) of the 5
wastewater and method of wastewater processing. Some resources such as consumables and 6
refrigerant use are relatively small and typically constitute a minimal proportion of the total GHG 7
emissions (e.g. <1%). Thus, secondary data on use of these resources are acceptable. 8
Potentially, a system expansion approach could be used whereby product displaced by the various 9
non-meat products could be determined. However, there are no published studies to guide such an 10
analysis. In view of the range of products and their usefulness for various purposes, it is recommended 11
to consider a system expansion approach in future when more relevant published research is available. 12
The PCR (Boeri, 2013) recommended the use of economic allocation. Similarly, the present guidelines 13
recommend the use of economic allocation. However, it is noted that some co-products may be 14
identified as of no economic value, but may be collected and used for secondary processing (e.g. used 15
for burning for energy or for producing blood-and-bone meal). 16
An example of the effects of economic allocation compared to mass allocation for the average lamb 17
from New Zealand abattoirs in mid-2009 (see Box 4) shows a much higher allocation to meat using 18
economic allocation than when mass allocation was used. Box 4 also shows the large variation in price 19
per kilogramme between different cuts of meat. The present guidelines recommend treating all meat 20
components as the same per-kg (i.e. no separate economic allocation between meat cuts). 21
Some abattoirs process multiple animal species (e.g. cattle and sheep). In such cases there is a need to 22
allocate emissions according to species. This shall be based on the relative number and live-weights of 23
the animal species processed. However, this approach will need to account for relative differences in 24
requirements (such as for energy use) between species; for example, the energy use per kg live-weight 25
processed for sheep can be about 1.3 to 2 times that for cattle. Similarly, some abattoirs may have an 26
associated rendering plant, and if energy use cannot be apportioned between meat processing and 27
rendering, some adjustment may be appropriate to account for the greater energy requirements for 28
rendering (e.g. associated with steam production). One available method is to apply specific energy-29
adjusted values based on survey data, where specific energy uses between rendering and non-30
rendering facilities have been obtained. For example, Lovatt and Kemp (1995) obtained specific fuel 31
use/tonne meat processed at eight-fold and two-fold higher for fuel and electricity use, respectively.32
76
BOX 4 EXAMPLE OF VARIATION IN THE MASS AND ECONOMIC VALUE OF COMPONENTS OF AN AVERAGE NEW 1 ZEALAND LAMB LEAVING AN ABATTOIR EFFECTS ON ALLOCATION CALCULATIONS (DATA PROVIDED BY NZ MEAT 2 INDUSTRY ASSOCIATION 2009) 3
4
Data in the table below were based on a summary of the average weight of different meat cuts and co-products 5 from lamb leaving an average abattoir in New Zealand in mid-2009. The associated average relative economic 6 value of the different components is also given and this is used to calculate the average allocation to meat. The 7 weighted average value across all edible components was used, thereby assuming no difference in “value” 8 between the different edible components when applying economic allocation. The table shows that in practice 9 there was more than an eight-fold difference in price per kilogramme between the lamb rack and neck cuts of 10 meat. It also illustrates the relatively large difference in economic value of the co-products. 11
12
Average mass of
component (kg)
Component % of
total mass
Price per‐kg relative to
leg meat
Component as % of
total economic value
Meat:
Neck 0.54 1.5 0.21 0.8
Shoulder 4.6 12.7 0.51 16.1
Rack 1.21 3.4 1.73 14.3
Breast and shank 1.46 4.1 0.47 4.8
Loin 1.43 4.0 1.04 10.2
Legs 4.68 13.0 1.00 32.1
Other meat 2.43 6.7 0.38 6.4
Edible offal 2.0 5.5 0.28 3.9
Co‐products:
Hide/skin 2.21 6.1 0.28 4.3
Wool 1.59 4.4 0.27 3.0
Blood 1.76 4.9 0.01 0.1
Inedible offal 0.65 1.8 0.14 0.6
Rendering/tallow 11.54 32.0 0.04 3.5
13
Thus the economic allocation percentage (EA) for meat relative to the total returns for the lamb was calculated 14 using: 15
EA (%) = 100 x Σ (weight of meat component i x relative value of meat component i) / [Σ (weight of meat 16 component i x relative value of meat component i) + Σ (weight of co-product i x relative value of co-product i)] 17
The mass allocation percentage (MA) for meat was calculated using: 18
MA (%) = 100 x Σ (weight of meat component i) / [Σ(weight of meat component i + Σ (weight of co-product i)] 19
The results from these calculations for % allocation to meat using economic or mass allocation were 88 percent 20 and 51 percent, respectively. 21
77
b) Calculating GHG emissions from meat processing 1
Calculation of GHG emissions shall account for resource use, wastewater processing, animal wastes 2
and the associated GHG emission factors. Electricity and other sources of energy use shall account for 3
total embodied emissions relevant to the country where the primary processing occurs. Data on 4
wastewater quantity and composition are used with the GHG emission factors for the method of 5
wastewater processing (IPCC, 2006) to calculate GHG emissions from wastewater processing. In 6
meat-processing plants, wastewater will generally include excreta from animals held prior to 7
processing, the contents of the stomachs and intestines of slaughtered animals, and various wastes 8
(e.g. blood, if not collected for further processing). However, where these sources are not specifically 9
captured in wastewater systems they shall be estimated and GHG emissions from them accounted for 10
using the IPCC (2006) method for waste. Total GHG emissions shall be allocated between the various 11
co-products, as outlined in Section 9.3. 12
To assist in understanding the relative importance of the various contributors to meat processing in 13
abattoirs, Ledgard et al. (2011) found from a survey of NZ sheep meat-processing plants that the 14
relative contributors to GHG emissions from electricity for chilling/freezing was 18 percent, other 15
energy use (particularly for water use and heating) 47 percent, wastewater processing 26 percent and 16
other sources 9 percent. 17
18
11.6.4 ON‐SITE ENERGY GENERATION 19
In some primary processing plants, waste material may be used for on-site energy generation. This 20
may simply be used to displace energy requirements within the plant, in which case net energy use 21
would be used in the analysis. Where there is a surplus of energy generated within the primary 22
processing system and sold outside the system under study, the present guidelines recommend the use 23
of system expansion. This is in line with ISO 14044 (2006). 24
25
12 INTERPRETATION OF LCA RESULTS 26
Interpretation of the results of the study serves two purposes (European Commission, 2010): 27
At all steps of the LCA, the calculation approaches and data shall match the goals and quality 28
requirements of the study. In this sense, interpretation of results may inform an iterative improvement 29
of the assessment until all goals and requirements are met. 30
The second purpose of the interpretation is to develop conclusions and recommendations, e.g. in 31
support of environmental performance improvements. The interpretation entails three main elements 32
detailed in the following subsections: "Identification of important issues," “Characterizing 33
uncertainty” and "Conclusions, limitations and recommendations". 34
78
12.1 Identification of key issues 1
Identifying important issues encompasses the identification of most important impact categories and 2
life cycle stages, as well the sensitivity of results to methodological choices. 3
The first step is to determine the life cycle stage processes and elementary flows that contribute most 4
to the LCIA results, as well as the most relevant impact categories. To do this, a contribution analysis 5
shall be conducted. It quantifies the relative contribution of the different stages/categories/items to the 6
total result. Such contribution analysis can be useful for various interests, such as focusing data 7
collection or mitigation efforts on the most contributing processes. 8
Secondly, the extent to which methodological choices such as system boundaries, cut-off criteria, data 9
sources, and allocation choices affect the study outcomes shall be assessed, especially impact 10
categories and life cycle stages having the most important contribution. In addition, any explicit 11
exclusion of supply chain activities, including those that are excluded as a result of cut-off criteria, 12
shall be documented in the report. Tools that should be used to assess the robustness of the footprint 13
model include (European Commission, 2010): 14
Completeness checks: evaluate the LCI data to confirm that it is consistent with the defined 15
goals, scope, system boundaries, and quality criteria and that the cut-off criteria have been 16
met. This includes completeness of process (i.e. at each supply chain stage, the relevant 17
processes or emissions contributing to the impact have been included) and exchanges (i.e. all 18
significant energy or material inputs and their associated emissions have been included for 19
each process). 20
Sensitivity checks: assess the extent to which the results are determined by specific 21
methodological choices, and the impact of implementing alternative, defensible choices where 22
these are identifiable. This is particularly important with respect to allocation choices. It is 23
useful to structure sensitivity checks for each phase of the study: goal and scope definition, the 24
life cycle inventory model, and impact assessment. 25
Consistency checks: ensure that the principles, assumptions, methods and data have been 26
applied consistently with the goal and scope throughout the study. In particular, ensure that 27
the following are addressed: (i) the data quality along the life cycle of the product and across 28
production systems, (ii) the methodological choices (e.g. allocation methods) across 29
production systems and (iii) the application of the impact assessments steps with the goal and 30
scope. 31
32
12.2 Characterizing uncertainty 33
This section is related to Section 9, data quality. Several sources of uncertainty are present in LCA. 34
First is knowledge uncertainty which reflects limits of what is known about a given datum, and second 35
79
is process uncertainty which reflects the inherent variability of processes. We can reduce knowledge 1
uncertainty by collecting more data. We may reduce process uncertainty by breaking complex 2
systems into smaller parts or aggregations, but inherent variability cannot be eliminated completely. 3
Third, the characterization factors that are used to combine the large number of inventory emissions 4
into impacts also bring uncertainty into the estimation of impacts. In addition, there is bias introduced 5
if the LCI model is missing processes, or may have larger flows than actually present. 6
Variation and uncertainty of data should be estimated and reported. This is important because results 7
based on average data (i.e. the mean of several measurements from a given process – at a single or 8
multiple facilities) or using LCIA characterization factors with known variance do not reveal the 9
uncertainty in the reported mean value of the impact. Uncertainty may be estimated and 10
communicated quantitatively through a sensitivity and uncertainty analysis and/or qualitatively 11
through a discussion. Understanding the sources and magnitude of uncertainty in the results is critical 12
for assessing robustness of decisions that may be made based on the study results. When mitigation 13
action is proposed, knowledge of the sensitivity to, and uncertainty associated with the changes 14
proposed provides valuable information regarding decision robustness, as described in Table 4. At a 15
minimum, efforts to accurately characterize stochastic uncertainty and its impact on the robustness of 16
decisions should focus on those supply chain stages or emissions identified as significant in the impact 17
assessment and interpretation. Where reporting to third parties, this uncertainty analysis shall be 18
conducted and reported. 19
20
TABLE 4: GUIDE FOR DECISION ROBUSTNESS FROM SENSITIVITY AND UNCERTAINTY 21
22
12.2.1 MONTE CARLO ANALYSIS 23
In a Monte Carlo analysis, parameters (LCI) are considered as stochastic variables with specified 24
probability distributions, quantified as probability density functions (PDF). For a large number of 25
realizations, the Monte Carlo analysis creates an LCA model with one particular value from the PDFs 26
of every parameter and calculates the LCA results. The statistical properties of the sample of LCA 27
results across the range of realizations are then investigated. For normally distributed data, variance is 28
Sensitivity Uncertainty Robustness
High High Low
High Low High
Low High High
Low Low High
80
typically described in terms of an average and standard deviation. Some databases, notably EcoInvent, 1
use a lognormal PDF to describe the uncertainty. Some software tools (e.g. SimaPro, OpenLCA) allow 2
the use of Monte Carlo simulations to characterize the uncertainty in the reported impacts as affected 3
by the uncertainty in the input parameters of the analysis. 4
5
12.2.2 SENSITIVITY ANALYSIS 6
Choice-related uncertainties arise from methodological including modelling principles, system 7
boundaries and cut-off criteria, choice of footprint impact assessment methods, and other assumptions 8
related to time, technology, geography, etc. Unlike the LCI and characterization factors, they are not 9
amenable to statistical description, but the sensitivity of the results to these choice-related uncertainties 10
can be characterized through scenario assessments (e .g., comparing the footprint derived from 11
different allocation choices) and/or uncertainty analysis (e.g. Monte Carlo simulations). 12
In addition to choice-related sensitivity evaluation, the relative sensitivity of specific activities (LCI 13
datasets) measures the percentage change in impact arising from a known change in input parameter 14
(Hong et al., 2010). 15
16
12.2.3 NORMALIZATION 17
According to ISO 14044, normalization is an optional step in impact assessment. Normalization is a 18
process in which an impact associated with the functional unit is compared against an estimate of the 19
entire regional impacts in that category (Sleeswijk et al., 2008). For example, livestock supply chains 20
have been estimated to contribute 14.5% of global anthropogenic greenhouse gas emissions (Gerber 21
et al., 2013). Similar assessments can be made at regional or national scales, provided that a 22
reasonably complete inventory of all emissions in that region which contribute to the impact category 23
exists. Normalization provides an additional degree of insight into those impacts for which significant 24
improvement would result in a significant improvement for the region in question, and can help 25
decision-makers to focus on supply chain hotspots for which improvement will result in the greatest 26
overall environmental benefit. 27
28
12.3 Conclusions, Recommendations and Limitations 29
The final part of interpretation is to draw conclusions derived from the results, pose answers to the 30
questions raised in the goal and scope definition stage, and recommend appropriate actions to the 31
intended audience, within the context of the goal and scope, explicitly accounting for limitations to 32
robustness, uncertainty and applicability. 33
81
Conclusions derived from the study should summarize supply chain "hot spots" derived from the 1
contribution analysis and the improvement potential associated with possible management 2
interventions. Conclusions should be given in the strict context of the stated goal and scope of the 3
study, and any limitation of the goal and scope can be discussed a posteriori in the conclusions. 4
As required under ISO 14044:2006, if the study is intended to support comparative assertions (i.e. 5
claims asserting difference in the merits of products based the study results), then it is necessary to 6
fully consider whether differences in method or data quality used in the model of the compared 7
products impair the comparison. Any inconsistencies in functional units, system boundaries, data 8
quality, or impact assessment shall be evaluated and communicated. 9
Recommendations are based on the final conclusion of the LCA study. They shall be logical, 10
reasonable, plausible founded and strictly relate to the goal of the study. Recommendations shall be 11
given jointly with limitations in order to avoid their misinterpretation beyond the scope of the study. 12
13
12.4 Use and comparability of results 14
It is important to note that these guidelines refer only to a partial LCA and that where results are 15
required for products throughout the whole life cycle then it is necessary to link this analysis with 16
relevant methods for secondary processing through to consumption and waste stages (e.g. EPD 2012; 17
PAS 2395 2013Draft). Results from application of these guidelines cannot be used to represent the 18
whole life cycle of small ruminant products. However, they can be used to identify hot-spots in the 19
cradle-to-primary-processing stages (which are major contributors to emissions across the whole life 20
cycle) and assess potential GHG reduction strategies. In addition, the functional units recommended 21
are intermediary points in the supply chains for virtually all small ruminant sector products and 22
therefore will not be suitable for a full LCA. But they can provide valuable guidance to practitioners to 23
the point of divergence from the system into different types of products. 24
25
12.5 Good practice in reporting LCA results 26
The LCA results and interpretation shall be fully and accurately reported, without bias and consistent 27
with the goal and scope of the study. The type and format of the report should be appropriate to the 28
scale and objectives of the study and the language should be accurate and understandable by the 29
intended user so as to minimise the risk of misinterpretation. 30
The description of the data and method shall be included in the report in sufficient detail and 31
transparency to clearly show the scope, limitations and complexity of the analysis. The selected 32
allocation method used shall be documented and any variation from the recommendations in these 33
guidelines shall be justified. 34
82
The report should include an extensive discussion of the limitations related to accounting for a small 1
numbers of impact categories and outputs. This discussion should address: 2
Negative impacts on other (non GHG) environmental criteria; 3
Positive environmental impacts (e.g., on biodiversity, landscape, carbon sequestration); 4
Multifunctional outputs other than production (e.g., economic, social, nutrition); 5 If intended for the public domain, a communication plan shall be developed to establish accurate 6
communication that is adapted to the target audience and defensible. 7
8
12.6 Report elements and structure 9
The following elements should be included in the LCA report: 10
Executive summary typically targeting a non-technical audience (e.g. decision-makers), 11
including key elements of goal and scope of the system studied and the main results and 12
recommendations while clearly giving assumptions and limitations; 13
Identification of the LCA study, including name, date, responsible organization or researchers, 14
objectives of/reasons for the study and intended users; 15
Goal of the study: intended applications and targeted audience, methodology including 16
consistency with these guidelines; 17
Functional unit and reference flows, including overview of species, geographical location and 18
regional relevance of the study; 19
System boundary and unit stages (e.g. to farmgate and farmgate to primary processing gate); 20
Materiality criteria and cut-off thresholds; 21
Allocation method(s) and justification if different from the recommendations in these 22
guidelines; 23
Description of inventory data: representativeness, averaging periods (if used), and assessment 24
of quality of data; 25
Description of assumptions or value choices made for the production and processing systems, 26
with justification; 27
Feed intake and application of LEAP Animal Feed Guidelines, including description of 28
emissions and removals (if estimated) for LUC; 29
LCI modelling and calculating LCI results; 30
Results and interpretation of the study and conclusions; 31
Description of the limitations and any trade-offs; 32
If intended for the public domain the report should also state whether or not the study was 33
subject to independent third-party verification. 34
83
12.7 Critical review 1
Internal review and iterative improvement should be carried out for any LCA study. In addition, if the 2
results are intended to be released to the public, third-party verification and/or external critical review 3
shall be undertaken (and should be undertaken for internal studies) to ensure that: 4
the methods used to carry out the LCA are consistent with these guidelines and are 5
scientifically and technically valid; 6
the data and assumptions used are appropriate and reasonable; 7
interpretations take into account the complexities and limitations inherent in LCA studies for 8
on-farm and primary processing; 9
the report is transparent, free from bias and sufficient for the intended user(s). 10
The critical review shall be undertaken by an individual or panel with appropriate expertise, e.g. 11
suitably qualified reviewers from agricultural industry or government or non-government officers with 12
experience in the assessed supply chains and LCA. Independent reviewers are highly preferable. 13
The panel report and critical review statement and recommendations shall be included in the study 14
report if publicly available. 15
84
13 REFERENCES 1
Alexandratos N. & Bruinsma, J. 2012. World agriculture towards 2030/2050: the 2012 revision. ESA Working paper 2
Rome, FAO 3
Annett, R.W. & Carson, A.F. 2011. A comparison of growth and carcass characteristics of hill lambs finished on a 4
selection of forage-based diets. Proceedings of the British Society of Animal Science Annual Meeting, 5
Nottingham, UK. 6
Benoit, M. & Dakpo, H. 2012. Greenhouse gas emissions on French meat sheep farms; Analysis over the period 7
1987-2010. Proceedings of the conference on emission of gas and dust from livestock, 10–13 June 2012, Saint 8
Malo, France. Rennes, France, INRA. 9
Benoit, M. & Laignel, G. 2011, Analyse sur le long terme de systèmes d'élevage ovins allaitants en France. Quelles 10
trajectoires et quels facteurs de réussite économique? INRA Productions Animales, 24 : 211-219. 11
Besier, B., Jacobson. C., Woodgate, R. & Bell, K. 2010. Sheep health. In D.J. Cottie, ed. International sheep and wool 12
handbook, pp. 471–488. Nottingham, UK, Nottingham University Press, United Kingdom. 13
Bett, R.C., Kosgey, I.S., Bebe, B.O. & Kahi, A.K. 2007. Breeding goals for the Kenya Dual Purpose goat: I. Model 14
development and application to smallholder production systems. Tropical Animal Health and Production, 39: 15
477-492. 16
Bhatta, R., Enishi, O., Takusari, N., Higuchi, K., Nonaka, L. & Kurihara, M. 2008. Diet effects on methane production 17
by goats and a comparison between measurement methodologies. Journal of Agricultural Science, 146(6): 18
705–715. 19
Bier, J.M., Verbeek, J.R. & Lay, M.C. 2012. An eco-profile of thermoplastic protein derived from blood meal. Part 1: 20
allocation issues. International Journal of Life Cycle Assessment, 17: 208–219. 21
Boeri, F. 2013. Product category rules: Meat of mammals. UN CPC 2111 and 2113. Environmental Product 22
Declaration. Stockholm, EPD. 23
BSI. 2011. Specification for the measurement of the embodied greenhouse gas emissions in products and services. 24
PAS 2050-2. London, British Standards Institution. 25
BSI. 2013. DRAFT PAS 2395 Specification for the assessment of greenhouse gas (GHG) emissions from the whole 26
life cycle of textile products. Draft 2.0 for DPC. London, British Standards Institution. 27
CSIRO. 2007. Nutrient requirements of domesticated ruminants. (M. Freer, H. Dove & J.V. Nolan, eds.) Clayton 28
South, Australia, CSIRO Publishing. 29
DairyCo. 2010. Guidelines for the carbon footprinting of dairy products in the UK. Kenilworth, UK, DairyCo, 30
DairyNZ. 2012. Facts and figures for New Zealand dairy farmers: Version 2. Newstead, New Zealand, DairyNZ 31
(available at www.dairynz.co.nz/page/pageid/2145866931/Facts_and_Figures). 32
Devendra, C., Morton, J.F. & Rischkovsky, B. 2005 Livestock systems. In E. Owen, A. Kitalyi, N. Jayasuria & 33
T. Smith, eds. Livestock and wealth creation, pp. 28–52. Nottingham, UK, Nottingham University Press. 34
85
Eady, S., Carre, A. & Grant, T. 2012. Life cycle assessment modeling of complex agricultural systems with multiple 1
food and fibre co-products. Journal of Cleaner Production, 28: 143–149. 2
EBLEX. 2012. Down to Earth, the beef and sheep roadmap: Phase three. Kenilworth, UK, EBLEX. 3
EBLEX. 2013. A summary of UK carbon footprint calculators for beef and sheep enterprises. Kenilworth, UK, 4
EBLEX. 5
Edwards-Jones, G., Plassmann, K. & Harris, I.M. 2009. Carbon footprinting of lamb and beef production systems: 6
Insights from an empirical analysis of farms in Wales. Journal of Agricultural Science, 147: 707–719. 7
EPD. 2011. Draft product category rule on woven fabrics of combed wool or of combed fine animal hair, containing 8
85% or more by weight of wool or fine animal hair. Environmental Product Declaration. UN CPC 2653. 9
Version 0.2. Stockholm, EPD. 10
EPD. 2012. Draft product category rule on textile yarn and thread of natural fibres and man-made filaments or staple 11
fibres. Environmental Product Declaration. UN CPC 263-264. Draft VI. Stockholm, EPD. 12
European Commission - Joint Research Centre - Institute for Environment and Sustainability: International 13
Reference Life Cycle Data System (ILCD) Handbook - General guide for Life Cycle Assessment - Detailed 14
guidance. First edition March 2010. EUR 24708 EN. Luxembourg. Publications Office of the European 15
Union; 2010 16
FAOSTAT 2013. http://faostat.fao.org/. FAO, Rome. 17
Finnveden G, Hauschild MZ, Ekvall T, 2009. Recent developments in Life Cycle Assessment. J Environ Manage 18
91:1–21. doi: 10.1016/j.jenvman.2009.06.018 19
Flysjö, A. 2011. Potential for improving the carbon footprint of butter and blend products. Journal of Dairy Science, 20
94: 5 833–5 841. 21
Flysjö, A., Cederberg, C., Henriksson, M. & Ledgard, S. 2012. The interaction between milk and beef production and 22
emissions from land use change – critical considerations in life cycle assessment and carbon footprint studies of 23
milk. Journal of Cleaner Production, 28: 134–142. 24
Flysjö, A., Henriksson, M., Cederberg, C., Ledgard, S. & Englund, J-E. 2011. The impact of various parameters on the 25
carbon footprint of milk production in New Zealand and Sweden. Agricultural Systems, 104: 459–469. 26
Forster, P., Ramaswamy, V., Artaxo, P., Berntsen, T., Betts, R., Fahey, D.W., Haywood, J., Lean, J., Lowe, D.C., 27
Myhre, G., Nganga, J., Prinn, R., Raga, G., Schulz, M. & Van Dorland, R. 2007. Changes in atmospheric 28
constituents and in radiative forcing. In S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis, K.B., Averyt, 29
M. Tignor & H.L. Miller, eds. Climate Change 2007: The Physical Science Basis. Contribution of Working 30
Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, 31
United Kingdom and New York, Cambridge University Press. 32
Frischknecht, R. & Rebitzer, G. 2005. The ecoinvent database system: a comprehensive web-based LCA database. 33
Journal of Cleaner Production, 13: 1 337–1 343. 34
Gac, A., Ledgard, S., Lorinquer, E., Boyes, M. & Le Gall, A. 2012. Carbon footprint of sheep farms in France and 35
New Zealand and methodology analysis. In M.S. Corson & H.M.G. der Werf, eds. Proceedings of the 8th 36
86
International Conference on Life Cycle Assessment in the Agri-Food Sector (LCA Food 2012), 1–4 October 1
2012, Saint Malo, France, pp. 310–314. Rennes, France, INRA. 2
Gerber, P., Vellinga, T., Opio, C., Henderson, B. & Steinfeld, H. 2010. Greenhouse gas emissions from the dairy 3
sector: A life cycle assessment. Rome, FAO. 4
Gerber, P.J., Steinfeld, H., Henderson, B., Mottet, A., Opio, C., Dijkman, J., Falcucci, A. and Tempio, G. 2013. 5
Tackling climate change through livestock – A global assessment of emissions and mitigation opportunities. 6
Food and Agriculture Organization of the United Nations (FAO), Rome. 7
Goedkoop, M.J., Heijungs, R., Huijbregts, M., De Schryver, A., Struijs, J. & Van Zelm, R. 2009. ReCiPe 2008 A life 8
cycle impact assessment method which comprises harmonised category indicators at the midpoint and the 9
endpoint level; First edition Report I: Characterisation. The Hague, Ministerie van VROM (available at 10
www.lcia-recipe.net). 11
Guinée, J.B., Gorrée, M., Heijungs, R., Huppes, G., Kleijn, R., de Koning, A., van Oers, L., Wegener Sleeswijk, A., 12
Suh, S., Udo de Haes, H.A., de Bruijn, H., van Duin, R., Huijbregts, M.A.J., Lindeijer, E., Roorda, A.A.H., van 13
der Ven, B.L. & Weidema, B.P. 2002. Handbook on life cycle assessment: Operational guide to the ISO 14
standards. Leiden, the Netherlands, Kluwer Academic Publishers. 15
Hong, J., Shaked, S., Rosenbaum, R.K. & Jolliet, O. 2010. Analytical uncertainty propagation in life cycle inventory 16
and impact assessment: Application to an automobile front panel. International Journal of Life Cycle 17
Assessment, 15: 499–510. 18
Huijbreghts, M.A.J., Rombouts, L.J., Hellweg, S., Frischknecht, R., Hendriks, A.J., van de Meent, D., Ragas, A.M.J., 19
Reijnders, L. & Struijs, J. 2006. Is cumulative fossil energy demand a useful indicator for the environmental 20
performance of products? Environment Science and Technology, 40(3): 641–648. 21
IDF. 2010a. A common carbon footprint approach for dairy: the IDF guide to standard lifecycle assessment 22
methodology for the dairy sector. Bulletin of the International Dairy Federation, 445/2010. 23
IDF. 2010b. IDF guide to life cycle assessment and life cycle management: A common carbon footprint methodology 24
for the global dairy sector. International Dairy Federation. Brussels, IDF (available at www.idf-lca-25
guide.org/Public/en/LCA+Guide/LCA+Guidelines+overview). 26
ILCD, 2010. ILCD Handbook: Analysis of existing Environmental Impact Assessment methodologies for use in Life 27
Cycle Assessment. European Commission Joint Research Centre, European Union. 115p. 28
IPCC. 2006. IPCC guidelines for national greenhouse gas inventories: Volume 4: Agriculture, forestry and other land 29
use. Intergovernmental Panel on Climate Change. Paris, IPCC (available at: www.ipcc-30
nggip.iges.or.jp/public/2006gl/vol4.htm). 31
ISO 14044. 2006. Environmental management – Life cycle assessment – Requirements and guidelines. 32
ISO 14044:2006(E). London, British Standards Institution. 33
ISO/TS 14067. 2013. Greenhouse gases: Carbon footprint of products – Requirements and guidelines for 34
quantification and communication. ISO/TC 207/SC 7. Geneva, ISO. 35
Joseph, K. & Nithya, N. 2009. Material flows in the life cycle of leather. Journal of Cleaner Production, 17: 676–682. 36
87
Lassey, K.R. 2007. Livestock methane emission: From the individual grazing animal through national inventories to 1
the global methane cycle. Agricultural and Forest Meteorology, 142:120–132. 2
Ledgard, S.F. 2011. A common carbon footprinting methodology for the lamb meat sector. (With contributions from 3
England, France, Northern Ireland, Scotland and Wales). Wellington, Beef + LambNZ / International Meat 4
Secretariat. 5
Ledgard, S.F., Lieffering, M., Coup, D. & O’Brien, B. 2011. Carbon footprinting of New Zealand lamb from an 6
exporting nations perspective. Animal Frontiers, 1: 27–32. 7
Ledgard S F, McDevitt J, Boyes M, Lieffering M & Kemp R. 2009. Greenhouse gas footprint of lamb meat: 8
Methodology report. Report for MAF. Hamilton, New Zealand, AgResearch. 9
Lieffering, M., Ledgard, S., Boyes, M. & Kemp, R. 2011. Venison greenhouse gas footprint: Final report. Report for 10
MAF. Hamilton, New Zealand, AgResearch. 11
Lovatt, S.J. & Kemp, R.M. 1995. Survey of utility use in the New Zealand meat industry for the 1993/94 season. 12
Technical report No. 952. Wellington, Meat Industry Research Institute of New Zealand (MIRINZ). 13
Mila, L., Domenech, X., Rieradevall, J., Fullana, P. & Puig, R. 1998. Application of life cycle assessment to footwear. 14
International Journal of Life Cycle Assessment, 3: 203–208. 15
Mila i Canals, L., Domenech, X., Rieradevall, J., Puig, R. & Fullana, P. 2002. Use of life cycle assessment in the 16
procedure for the establishment of environmental criteria in the Catalan eco-label of leather. International 17
Journal of Life Cycle Assessment, 7: 39–46. 18
Muir, P.D., Thomson, B.C. & Askin, D.C. 2008. A review of dressing out percentage in New Zealand livestock. 19
Report for MAF. Wellington, Ministry of Agriculture and Forestry. 20
NRC. 2007. Nutrient requirements of small ruminants: Sheep, goats, cervids, and new world camelids. National 21
Research Council. Washington DC, National Academies Press. 22
Opio, C., Gerber, P., MacLeod, B., Falcucci, A., Henderson, B., Mottet, A. Tempio, G., & Steinfeld, H. 2013. 23
Greenhouse gas emissions from ruminant supply chains: A global life cycle assessment. Rome, FAO. 24
Park, Y.W. & Haenlein, G.F.W. 2006. Handbook of milk of non-bovine mammals. Oxford, Blackwell Publishing. 25
Peters, G.M., Rowley, H.V., Wiedemann, S., Tucker, R., Short, M.D. & Schulz, M. 2010. Red meat production in 26
Australia: Life cycle assessment and comparison with overseas studies. Environmental Science and 27
Technology, 44: 1327–1332. 28
Petersen, A.K. & Solberg, B. 2004. Greenhouse gas emissions and costs over the life cycle of wood and alternative 29
flooring materials. Climate Change, 64: 143–167. 30
Pickering, A. 2011. Detailed methodologies for agricultural greenhouse gas emission calculation. MAF Technical 31
Paper No. 2011/40. Wellington, Ministry of Agriculture and Forestry. 32
Potting, J. & Blok, K. 1995. Life-cycle assessment of four types of floor covering. Journal of Cleaner Production, 3: 33
201–213. 34
88
Pulina, G., Macciotta, N & Nudda, A. 2005. Milk composition and feeding in the Italian dairy sheep. Italian Journal of 1
Animal Science, 4 (Suppl. 1): 5-14. 2
Ramirez, A.D., Humphries, A.C., Woodgate, S.L. & Wilkinson, R.G. 2011. Greenhouse gas life cycle assessment of 3
products arising from the rendering of mammalian animal byproducts in the UK. Environmental Science and 4
Technology, 46: 447–453. 5
Ridoutt, B.G. & Pfister, S. 2010. A revised approach to water footprinting to make transparent the impacts of 6
consumption and production on global freshwater scarcity. Global Environmental Change, 20: 113–120. 7
Ripoll-Bosch, R., de Boer, I.J.M., Bernués, A. & Vellinga, T.V. 2013. Accounting for multi-functionality of sheep 8
farming in the carbon footprint of lamb: A comparison of three contrasting Mediterranean systems. 9
Agricultural Systems, 116: 60–68. 10
Rossi, S. 2012. Product category rules: Textile yarn and thread of natural fibres and man-made filaments or staple 11
fibres. Draft VI CPC Division 263-264. Environmental Product Declaration. Stockholm, EPD. 12
Sessa, F. 2013a. Product category rules: Processed liquid milk and cream. CPC 221. Environmental Product 13
Declaration. Stockholm, EPD. 14
Sessa, F. 2013b. Product category rules: Yoghurt, butter and cheese. CPC 2223, 2224 and 2225. Environmental 15
Product Declaration. Stockholm, EPD. 16
Sleeswijk, A. W., van Oers, L. F., Guinée, J. B., Struijs, J., & Huijbregts, M. A. 2008. Normalisation in product life 17
cycle assessment: An LCA of the global and European economic systems in the year 2000. Science of the total 18
environment, 390(1), 227-240. 19
Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M. & de Haan, C. 2006. Livestock’s long shadow: 20
Environmental issues and options. Rome, FAO. 21
Swainson, N.M, Hoskin, S.O., Clark, H., Pinares-Patino, C.S. & Brooks, I.M. 2008. Comparative methane production 22
and yields from adult cattle, red deer and sheep. Australian Journal of Experimental Agriculture, 48(1-2): 23
lxxix–lxxx. 24
Thamsiriroj, T. & Murphy, J.D. 2011. The impact of the life cycle analysis methodology on whether biodiesel 25
produced from residues can meet the EU sustainability criteria for biofuel facilities constructed after 2017. 26
Renewable Energy, 36: 50–63. 27
Thoma, G., Jolliet, O. & Wang, Y. 2013a. A biophysical approach to allocation of life cycle environmental burdens for 28
fluid milk supply chain analysis. International Dairy Journal, 31: S41–S49. 29
Thoma, G., Popp, J., Nutter, D., Shonnard, D., Ulrich, R., Matlock, M., Kim, D.S., Neiderman, Z., Kemper, N., East, 30
C. & Adom, F. 2013b. Greenhouse gas emissions from milk production and consumption in the United States: 31
A cradle-to-grave life cycle assessment circa 2008. International Dairy Journal, 31: S3–S14. 32
Wallman, M., Cederberg, C. & Sonesson, U. 2012. Life cycle assessment of Swedish lamb production. Göteborg, 33
Sweden, Swedish Institute for Food and Biotechnology (SIK). 34
89
Williams, A., Pell, E., Webb, J., Moorhouse, E. & Audsley, E. 2008. Comparative life cycle assessment of food 1
commodities procured for UK consumption through a diversity of supply chains. DEFRA Project FO0103. 2
London, Department for Environment, Food and Rural Affairs. 3
Witney, B. 1988. Choosing and using farm machinery. Harlow, UK, Longman Scientific and Technical. 4
WRI/WBCSD. 2011. Greenhouse gas protocol: Product life cycle accounting and reporting standard. 5
Washington DC, World Resources Institute and World Business Council for Sustainable Development.6
90
APPENDICES 1
91
Appendix 1: Greenhouse gas emissions from small ruminants: A review of 1
existing methodologies and guidelines 2
3
Introduction 4
Greenhouse gas (GHG) emissions from livestock systems have been identified as a significant 5
contributor to total global emissions (e.g. Steinfeld et al., 2006). This was defined as being of 6
particular significance for ruminant animals because of their high enteric methane emissions. 7
There have been many published studies of GHG emissions from livestock systems globally. 8
However, the methodologies used for estimating GHG emissions have varied widely. Various authors 9
have highlighted the difficulties in making comparisons across published studies because of the large 10
differences in methodologies used (e.g. Edwards-Jones et al., 2009; Flysjö et al., 2011). Consequently, 11
there has been interest in trying to agree on a common methodology for estimating GHG emissions 12
both between and within sectors. In 2010, the International Dairy Federation (IDF, 2010b) developed a 13
common methodology for estimating the carbon footprint (i.e. total GHG emissions) for dairy 14
products. Estimates of total GHG emissions are now often been based on use of life cycle assessment 15
(LCA) to account for all GHG sources and to determine the extent of emissions on a product basis. 16
The purpose of this brief review is to summarize existing methodologies and guidelines for calculating 17
GHG emissions from small ruminants. 18
19
GHG methodology guidelines for small ruminants 20
There have been two international methodology reports relating to small ruminants or sheep 21
specifically. Gerber et al. (2013) published a report on the carbon footprint of livestock that included 22
beef, sheep and goats for a range of agro-ecology regions and production systems around the world. 23
This was based on methodology developed by staff in the FAO and covered the cradle-to-farm-gate, 24
meat-processing and transportation-to-retailer stages of the life cycle. 25
Beef + Lamb New Zealand and the International Meat Secretariat initiated development of a “straw-26
man” document entitled A common carbon footprint methodology for the lamb meat sector (Ledgard, 27
2011) that involved contributions from various international industry and sheep LCA research groups, 28
including the English Beef and Lamb Executive (EBLEX), Adrian Williams (Cranfield University, 29
UK), Ronald Annett (AFBI, Northern Ireland), the Institut de l’Elevage (France), Hybu Cig Cymru 30
(HCC Wales) and Quality Meats Scotland. This document was confined to the cradle-to-farm-gate 31
stage of the lamb life cycle. It followed a similar approach to that of the IDF (2010b) common 32
methodology for milk production, with the intention of including a number of recommendations on 33
92
methodology aspects where specific methodology choices are required (e.g. system boundary, 1
functional unit and allocation methods). 2
Only a few publications or reports have estimated the total GHG emissions from sheep production, 3
and only one study for deer (Lieffering et al., 2011) and one for goats (Gerber et al., 2013) could be 4
found via a detailed literature search. Table 5 provides a summary of the published sheep studies with 5
carbon footprint estimates for lamb and the variation in components of the methodology used. Studies 6
varied in the extent of their system boundary and therefore in the relevant functional unit. 7
93
TABLE 5: SUMMARY OF THE CARBON FOOTPRINT OF LAMB FROM PUBLISHED STUDIES AND METHODOLOGIES USED 1
Reference Data source System boundary Functional unit (FU) Allocation method(s) Enteric methane
Carbon footprint
(kg CO2‐eq/FU)
International:
Ruminants: Gerber et al. (2013) FAO country data RDC1 1 kg CW2,
1 kg FPCM3
Economic, protein content Tier 2 15‐314
(4.7‐9.05)
Lamb: Ledgard (2011) Representative data Farm gate 1 kg LW Biophysical/economic Tier 2 n.a.6
Country‐specific:
Australia: Peters et al. (2010) 1 case farm Farm gate 1 kg CW Mass Tier 2 10.5
Australia: Eady et al. (2012) 1 case farm Farm gate 1 kg CW System expansion/ biophysical/economic
Tier 2 12.6
England: EBLEX (2012) 57 case farms Farm gate 1 kg LW Economic Tier 2 6‐207
France: Gac et al. (2012) Farm survey (104)8/model Farm gate 1 kg LW Mass Tier 1 12.9
France: Benoit and Dakpo (2012) Farm survey (1180) Farm gate 1 kg CW Mass Tiers 1 and 2 11.9 (15‐82)9
NZ: Ledgard et al. (2011) Farm survey (437)/model Life cycle10 1 kg meat Biophysical/economic Tier 2 19
Spain: Ripoll‐Bosch et al. (2013) Farm survey (3) Farm gate 1 kg LW Economic Tier 2 19‐26
Sweden: Wallman et al. (2012) 10 case farms RDC 1 kg CW Mass/economic Tier 2 16
UK: Williams et al. (2008) UK model RDC 1 kg CW Economic Tier 2 14.1
Wales: Edwards‐Jones et al. (2009) 2 case farms Farm gate 1 kg LW Economic Tier 1 8‐14411
Note: 1 to retail distribution centre; 2 carcass-weight; 3 fat and protein corrected milk; 4 range for small ruminant CW across regions globally; 5 range for small ruminant FPCM across regions globally; 2 6 generic methodology only (no specific analyses included); 7 average and range across 57 case farms; 8 bracketed values refer to number of farms in surveys; 9 range across 1 180 farms (or -7 to 62 if 3 carbon sequestration was included); 10 whole life cycle (i.e. cradle-to-grave); and 11 high on peat soils. 4
94
The study of Eady et al. (2012) was for a case farm with mixed cropping and livestock. The authors 1
used system expansion to allocate between crop and livestock and compared biophysical and 2
economic allocation for lamb/mutton/wool. Similarly, the NZ system (Ledgard et al., 2011) included 3
mixed sheep and cattle farming and used biophysical allocation to allocate between each animal type 4
(i.e. apportioning according to the amount of feed dry matter consumed), and then used economic 5
allocation for lamb/mutton/wool. Enteric methane was a significant contributor to the carbon footprint 6
and therefore most studies used a tier-2 methodology, whereby feed intake was estimated from a 7
number of animal productivity parameters (e.g. live-weight, growth rate, lambing percentage and 8
replacement rate). However, two studies used a tier-1 methodology where each sheep class had a 9
constant methane emission per animal. In view of the large contribution from enteric methane, it is 10
desirable to use a tier-2 methodology since there can be large differences in animal productivity, feed 11
conversion efficiency and methane emissions per kg animal production, including from sheep 12
(e.g. Ledgard et al., 2011; Benoit and Dakpo, 2012). 13
There is only one published carbon footprint study for goats, which showed similar or slightly lower 14
values than from lamb or sheep meat (Gerber et al., 2013). The principles for estimating the carbon 15
footprint for goat products are likely to be similar to those for sheep. Indeed, a specific study on 16
enteric methane emissions by goats in respiration chambers on different diets showed an average of 17
6.6 percent of energy loss as methane (Bhatta et al., 2008). This is similar to that from other animal 18
studies and to the IPCC default value of 6.5 percent. Nevertheless, comparative animal studies on 19
enteric methane emissions indicated significant differences between sheep and deer fed the same diet, 20
with average values of 18.4 and 16.5 g methane/kg dry matter intake, respectively (Swainson et al., 21
2008). 22
In conclusion, the estimates of the carbon footprint of lamb shown in Table 5 showed wide variability 23
and much of this variability can be attributed to differences in methodology used. This highlights the 24
importance of developing a common methodology to be able to identify real differences in GHG 25
emissions between farm systems and products, and to identify GHG reduction opportunities. 26
27
Product category rules 28
The generic GHG methodology guidelines refer to product category rules (PCRs) and recommend that 29
these are used where they have been produced. A detailed search revealed that there are no specific 30
PCRs for sheep or goat products. However, there are generic PCRs on “Meat of mammals” (Boeri, 31
2013), “Processed liquid milk” (Sessa, 2013a) and a draft PCR on “Textile yarn and thread from 32
natural fibres, man-made filaments or staple fibres” (Rossi, 2012), which can be used to assist in 33
developing methodology guidelines for small ruminants. 34
95
GHG footprinting tools covering small ruminants 1
There are a number of GHG footprinting tools that are being used or available for use on farms for 2
evaluation of the GHG footprint of small ruminants and mitigation options. Ten carbon calculators 3
available within the United Kingdom were recently reviewed by EBLEX (2013). Many of these use an 4
LCA approach and account for UK-specific management practices, but in most cases the specific 5
methodology and algorithms are not published and therefore specific methodology details are 6
unavailable. This makes it difficult to assess these models, but it gives an indication of the potential 7
for practical use on-farm. It also highlights the importance in having a commonly agreed methodology 8
for estimating GHG emissions from small ruminants and their products for comparison of production 9
and processing systems. 10
11
References 12
Benoit, M. & Dakpo, H. 2012. Greenhouse gas emissions on French meat sheep farms; Analysis over the period 13
1987-2010. Proceedings of the conference on emission of gas and dust from livestock, 10–13 June 2012, Saint 14
Malo, France. Rennes, France, INRA. 15
Bhatta, R., Enishi, O., Takusari, N., Higuchi, K., Nonaka, L. & Kurihara, M. 2008. Diet effects on methane production 16
by goats and a comparison between measurement methodologies. Journal of Agricultural Science, 146(6): 17
705–715. 18
BSI. 2011. Specification for the measurement of the embodied greenhouse gas emissions in products and services. 19
PAS 2050-2. London, British Standards Institution. 20
Eady, S., Carre, A. & Grant, T. 2012. Life cycle assessment modeling of complex agricultural systems with multiple 21
food and fibre co-products. Journal of Cleaner Production, 28: 143–149. 22
EBLEX. 2012. Down to Earth, the beef and sheep roadmap: Phase three. Kenilworth, UK, EBLEX. 23
EBLEX. 2013. A summary of UK carbon footprint calculators for beef and sheep enterprises. Kenilworth, UK, 24
EBLEX. 25
Edwards-Jones, G., Plassmann, K. & Harris, I.M. 2009. Carbon footprinting of lamb and beef production systems: 26
Insights from an empirical analysis of farms in Wales. Journal of Agricultural Science, 147: 707–719. 27
Flysjö, A., Henriksson, M., Cederberg, C., Ledgard, S. & Englund, J-E. 2011. The impact of various parameters on the 28
carbon footprint of milk production in New Zealand and Sweden. Agricultural Systems, 104: 459–469. 29
Gac, A., Ledgard, S., Lorinquer, E., Boyes, M. & Le Gall, A. 2012. Carbon footprint of sheep farms in France and 30
New Zealand and methodology analysis. In M.S. Corson & H.M.G. der Werf, eds. Proceedings of the 8th 31
International Conference on Life Cycle Assessment in the Agri-Food Sector (LCA Food 2012), 1–4 October 32
2012, Saint Malo, France, pp. 310–314. Rennes, France, INRA. 33
Gerber, P., Vellinga, T., Opio, C., Henderson, B. & Steinfeld, H. 2010. Greenhouse gas emissions from the dairy 34
sector: A life cycle assessment. Rome, FAO. 35
96
IDF. 2010b. IDF guide to life cycle assessment and life cycle management: A common carbon footprint methodology 1
for the global dairy sector. International Dairy Federation. Brussels, IDF (available at www.idf-lca-2
guide.org/Public/en/LCA+Guide/LCA+Guidelines+overview). 3
Ledgard, S.F. 2011. A common carbon footprinting methodology for the lamb meat sector. (With contributions from 4
England, France, Northern Ireland, Scotland and Wales). Wellington, Beef + LambNZ / International Meat 5
Secretariat. 6
Ledgard, S.F., Lieffering, M., Coup, D. & O’Brien, B. 2011. Carbon footprinting of New Zealand lamb from an 7
exporting nations perspective. Animal Frontiers, 1: 27–32. 8
Ledgard S F, McDevitt J, Boyes M, Lieffering M & Kemp R. 2009. Greenhouse gas footprint of lamb meat: 9
Methodology report. Report for MAF. Hamilton, New Zealand, AgResearch. 10
Lieffering, M., Ledgard, S., Boyes, M. & Kemp, R. 2011. Venison greenhouse gas footprint: Final report. Report for 11
MAF. Hamilton, New Zealand, AgResearch. 12
Nguyen, T.L.T., Hermansen, J.E. & Mogensen, L. 2010. Environmental consequences of different beef production 13
systems in the EU. Journal of Cleaner Production, 18: 756–766. 14
Peters, G.M., Rowley, H.V., Wiedemann, S., Tucker, R., Short, M.D. & Schulz, M. 2010. Red meat production in 15
Australia: Life cycle assessment and comparison with overseas studies. Environmental Science and 16
Technology, 44: 1327–1332. 17
Ripoll-Bosch, R., de Boer, I.J.M., Bernués, A. & Vellinga, T.V. 2013. Accounting for multi-functionality of sheep 18
farming in the carbon footprint of lamb: A comparison of three contrasting Mediterranean systems. 19
Agricultural Systems, 116: 60–68. 20
Steinfeld, H., Gerber, P., Wassenaar, T., Castel, V., Rosales, M. & de Haan, C. 2006. Livestock’s long shadow: 21
Environmental issues and options. Rome, FAO. 22
Swainson, N.M, Hoskin, S.O., Clark, H., Pinares-Patino, C.S. & Brooks, I.M. 2008. Comparative methane production 23
and yields from adult cattle, red deer and sheep. Australian Journal of Experimental Agriculture, 48(1-2): 24
lxxix–lxxx. 25
Wallman, M., Cederberg, C. & Sonesson, U. 2012. Life cycle assessment of Swedish lamb production. Göteborg, 26
Sweden, Swedish Institute for Food and Biotechnology (SIK). 27
Williams, A., Pell, E., Webb, J., Moorhouse, E. & Audsley, E. 2008. Comparative life cycle assessment of food 28
commodities procured for UK consumption through a diversity of supply chains. DEFRA Project FO0103. 29
London, Department for Environment, Food and Rural Affairs. 30
97
Appendix 2: Small ruminants: Main producing countries 1
2
TABLE 6: RELATIVE NUMBER OF GOATS (IN GRASSLAND VS. MIXED PRODUCTION SYSTEMS) AND OF GOAT MEAT 3 AND MILK PRODUCTION ACROSS COUNTRIES GLOBALLY, 2011 4
Country Goats (thousand head) Goat meat (thousand tonnes)
Goat milk (thousand tonnes) Grassland‐based
systems Mixed systems
Top 20 countries (for herd)
China 22178 129036 1890 277
India 2232 121298 599 4760
Pakistan 13404 42590 286 759
Nigeria 2159 47788 289 ‐
Bangladesh 0 44208 199 2496
Sudan (former) 20411 22113 97 1072
Iran (Islamic Republic of) 18447 7061 163 306
Ethiopia 2081 14283 68 53
Somalia 12820 1751 65 501
Kenya 6363 7510 47 197
Indonesia 320 12997 71 281
Mongolia 11133 2114 48 71
United Republic of Tanzania 2112 10424 35 108
Niger 7381 3857 66 278
Burkina Faso 1470 9239 35 106
Brazil 3637 6620 29 148
Mexico 1985 6884 44 162
Mali 3742 4963 75 703
Yemen 2232 5630 42 56
Uganda 161 7638 33 ‐
Remaining countries 66412 108655 1045 4758
Sources: Gerber et al. (2013) for grassland vs. mixed production systems and FAOSTAT (2013) for goat meat and milk 5 production. 6 7
98
TABLE 7: RELATIVE NUMBER OF SHEEP (IN GRASSLAND VS. MIXED PRODUCTION SYSTEMS) AND SHEEP MEAT, MILK 1 AND WOOL PRODUCTION ACROSS COUNTRIES GLOBALLY, 2011 2
Country Sheep (thousand head) Sheep meat (thousand tonnes)
Sheep milk (thousand tonnes)
Greasy wool (thousand tonnes)
Grassland‐based systems
Mixed systems
Top 20 countries (for herd)
China 68590 83025 2050 1529 393
Australia 47434 53667 564 ‐ 368
India 2138 60330 293 ‐ 43
Iran (Islamic Republic of) 36090 17648 96 449 60
Sudan (former) 23865 25929 215 390 55
New Zealand 12015 27864 465 ‐ 166
United Kingdom 4894 30066 301 ‐ 67
Nigeria 1235 30304 168 ‐ ‐
South Africa 16850 8291 124 ‐ 41
Turkey 5534 19536 230 893 47
Pakistan 6228 18522 158 36 42
Spain 1287 21230 135 520 22
Ethiopia 1635 19099 88 58 8
Syrian Arab Republic 11702 7882 183 706 21
Algeria 10268 8641 253 320 26
Morocco 4076 12793 144 38 55
Brazil 5847 9659 84 ‐ 12
Russian Federation 5233 10063 171 1 53
Peru 4858 9950 35 ‐ 10
Somalia 13117 1577 70 590 ‐
Remaining countries 127899 190774 2402 4241 553
Sources: Gerber et al. (2013) for grassland vs. mixed production systems and FAOSTAT (2013) for sheep meat, milk and 3 wool production. 4
99
Appendix 3: Summary of carcass‐weight: live‐weight ratios (as percentages) 1
for goats and sheep for different regions 2
3
TABLE 8: AVERAGE RATIOS OF CARCASS‐WEIGHT TO LIVE‐WEIGHT FOR GOATS AND SHEEP FOR DIFFERENT GLOBAL 4 REGIONS 5
Region Goats Sheep
North America 52 52
Russian Federation 43 45
Near East and North Africa 43 48
Western Europe 43 45
Eastern Europe 44 45
East and South East Asia 48 49
Oceania 45 50
South Asia 43 48
Latin American countries 44 49
Sub‐Saharan Africa 48 45
Source: Based on a summary by Gerber et al. (2013). 6 7
Carcass-weight (CW), sometimes called dead-weight, generally refers to the weight of the carcass 8
after removal of the skin, head, feet and internal organs including the digestive tract (and sometimes 9
some surplus fat). The “hot carcass-weight” may be recorded after slaughter and refers to the unit by 10
which farmers in some countries are paid. In practice, the carcass loses a small amount of moisture as 11
it cools (e.g. about 1-2 percent) to the cold CW. 12
The variation in these average default CW values of 43-52 percent for goats and 45-52 percent for 13
sheep probably reflects differences in method of calculation from the literature that it was derived 14
from (e.g. hot versus cold CW), as well as differences associated with key factors of age, breed, 15
weight, gender and diet. For example, in a review of New Zealand data, the hot-CW:LW ratio 16
averaged 44 percent (range 40-48 percent) for lambs and 39 percent for ewes/rams (Muir et al., 2008), 17
while for Northern Ireland it was 46 percent for lambs reared on pasture and 49 percent for lambs 18
reared on concentrates (Annett and Carlson, 2011). 19
100
Appendix 4: Allocation options for manure exported off‐farm 1
In some situations, part or all of the excreta from animals may be collected and sent off-farm. The 2
following is based on the decision tree shown in Figure 7. Unless a system separation approach is 3
adopted for manure, that is following flow 1a, or substitution (following flow 3a), a decision 4
associated with the classification of manure as either a co-product, waste or residual must be made in 5
box 3 of the decision tree. It is the strong recommendation of these guidelines that the default manure 6
method should be to consider it as a residual at the farm gate. This results in a clean separation of the 7
system where all post-farm emissions from use of the manure are assigned to that use, while all on-8
farm management is assigned to the main product from the farm (animal live-weight and fibre). 9
Substitution: When manure is used directly to produce energy, system expansion may be used to 10
unambiguously substitute for the energy displaced from the external economy. The primary question 11
will be whether marginal or average energy products are substituted.4 For the purposes of these 12
guidelines, if system expansion/substitution is used, then the average energy product (e.g. national 13
electricity grid mix of primary energy sources) shall be substituted. Combustion ash must also be 14
evaluated as a potential co-product, residual or waste, and associated emissions properly accounted 15
for. In addition, clear documentation of the substitution shall be included in the report and, following 16
other extant standards if energy is sold, then no credit as a substitute product is allowed. However, in 17
the latter case, the energy shall be considered as a co-product and an appropriate allocation chosen 18
through use of the decision tree. 19
Manure may be sold off-farm as a fertilizer replacement. This option of substitution as fertilizer shall 20
only be considered where clear evidence can be provided about specific replacement of fertilizer 21
products, and justification shall be provided to document the equivalence of the substitution. 22
Utilization of manure as fertilizer also results in different emissions from the field other than inorganic 23
fertilizers. Therefore, substitution shall require assignment of the field emissions to the animal 24
product, with a subsequent substitution credit of both the production and field emissions associated 25
with the substituted inorganic fertilizer. 26
Co-product: When manure is a valuable output of the farm, and if the system of manure production 27
cannot be separated from the system of animal production, then the full supply chain emissions to the 28
farm gate shall be shared by these two co-products. In this case, physical allocation based on physical 29
parameters is clearly not appropriate (since it is a completely different product to the other co-30
products, such as meat). Thus, an economic allocation shall be adopted. There shall be no credit for 31
4 Blonk et al. (2010) show that, in practice, this is not immediately obvious for an electricity grid. The grid is fed by multiple
production units and how should the avoided production/consumption mix be determined if your type of electricity
production is a part of the mix. BSI (2011) defines a practical approach by simply stating that the average country production
mix should be applied.
101
avoided fertilizer production in this situation, because the option for system substitution is path 3a, not 1
3c. As a co-product, it would have a functional unit, which would be 1 kg manure with a defined set of 2
characteristics (i.e. dry matter, digestibility and nitrogen concentrations). 3
Residual: This applies when manure has essentially no value as it leaves the farm gate. This is 4
equivalent to system separation, in that activities associated with conversion of the residual to a useful 5
product occur outside of the system boundary for the small ruminant production system. In such cases, 6
emissions linked to its production would go to the other co-product(s) (including emissions on 7
storage), while subsequent emissions from transport and use (e.g. on a crop) would go to that 8
subsequent use (i.e. outside the system boundary of these guidelines). 9
Waste: If manure is classified as waste-only in situations where it is disposed by landfill, incineration 10
without energy recovery or sent to a treatment facility, then all on-farm emissions shall be assigned to 11
the animal products of live-weight and greasy fibre. In such cases, emissions associated with the final 12
disposition of manure are within the system boundary and must be accounted for and assigned to the 13
animal products. In the case of manure as a waste, there shall be no substitution credit for its use as 14
fertilizer since this falls under path 3a, not 3e. 15
16
References 17
Blonk, H., Kool, A., Luske, B., Ponsioen, T., & Scholten, J. 2010. Methodology for assessing carbon footprints 18
of horticultural products. A study of methodological issues and solutions for the development of the 19
Dutch carbon footprint protocol for horticultural products. Blonk Milieu Advies, Gouda, the Netherlands 20
102
Appendix 5: Average sheep flock and goat herd parameters for different regions of the world 1
Parameters North America Russian
Federation Western Europe
Eastern Europe
North Africa and Near
East
East and South East
Asia Oceania South Asia
Latin America and the
Caribbean
Sub‐Saharan Africa
Sheep: weights (kg)
Adult female 80 49 62 44 41 47 70 35 59 38
Adult male 108 101 82 85 55 65 98 45 81 51
Lambs at birth 4 3 4 3 3 4 4 3 3 3
Slaughter female 27 21 29 21 26 26 35 24 29 24
Slaughter male 27 21 29 21 26 26 35 24 29 24
Rates (%)
Replacement female 21 23 29 22 21 16 24 18 20 17
Fertility 92 95 91 90 83 77 100 81 91 76
Death rate ‐ lambs 19 17 18 18 25 31 9 24 18 33
Death rate ‐ other 8 2 3 5 12 14 4 12 12 13
Age at first lambing (years) 2.1 1.9 1.6 1.8 1.4 1.6 1.8 1.6 2 1.5
Goats: weights (kg)
Adult female 64 55 59 (61)* 50 37 (40) 44 (34) 50 32 (31) 35 (37) 29 (31)
Adult male 83 100 88 (91) 100 53 (56) 53 (43) 81 42 (39) 50 (60) 36 40)
Kids at birth 6.4 2.2 4.0 (4.6) 5 2.7 (3.2) 3.9 (2.1) 3.6 2.7 (2.4) 3.5 (3.7) 2.2 (2.3)
Slaughter female 36 30 26 30 32 27 38 25 27 19
Slaughter male 36 30 26 30 32 27 38 25 28 19
Rates (%)
Replacement female 30 18 17 18 19 24 21 19 24 16
Fertility 85 90 87 90 87 88 87 81 80 87
Death rate ‐ kids 18 5 4 5 31 37 12 15 14 27
Death rate ‐ other 9 2 2 2 7 16 6 5 5 7
Age at first kidding (years) 1.4 1.3 1.3 1.3 1.6 1.1 1.4 1.8 1.5 2
Note: * Numbers in brackets refer to the parameters for meat animals. Source: Gerber et al. (2011). 2
103
Appendix 6: Calculation of enteric methane emissions from animal energy 1
requirements 2
3
Background 4
Section 11.2.2.b outlines the procedure for calculating the energy requirements for animal class 5
(e.g. lambs, hoggets, ewes, rams for sheep) according to metabolic requirements for the relevant 6
categories of growth, maintenance, activity (walking), reproduction, wool production and milk 7
production. These calculations are based on net energy (NE) as used in IPCC (2006) or metabolizable 8
energy (ME) intake. 9
However, the procedures for calculating enteric methane are usually described as a percentage of 10
gross energy (GE) intake. Thus, there is a need to convert NE or ME to GE. shows the relationship 11
between these, where GE can be partitioned to manure energy and enteric methane energy and net 12
energy. 13
14
FIGURE 13: DIAGRAM SHOWING THE FLOW OF THE DIFFERENT SOURCES OF ENERGY FOR RUMINANTS, BASED ON 15 A HIGH‐QUALITY FEED WITH A DIGESTIBILITY OF 75% 16
17
Source: Lassey (2007). 18 19
Calculation of gross energy 20
The main additional data needed are the % digestibility of the feed. A summary of the range of values 21
for different feed types is given later in this Appendix. 22
104
IPCC (2006) uses net energy and gives the following equation for the ratio of NE for growth to the 1
digestible energy consumed (REG): 2
3
REG = [1.164 – (5.160 x 103 x DE%) + (1.038 x 10-5 x DE%2) - (37.4/DE%)] 4
where DE% is digestible energy as a % of gross energy in the feed. 5
6
Similarly, the following equation is used for the ratio of net energy for maintenance to the digestible 7
energy consumed (REM): 8
9
REM = [1.123 – (4.092 x 103 x DE%) + (1.126 x 10-5 x DE%2) - (25.4/DE%)]. 10
11
From these, the gross energy (GE in MJ/day) is calculated using: 12
13
GE = [NEm +NEa +NEl + NEr] + [NEg + New] ÷ (DE%/100) 14
REM REG 15
16
where the subscripts m, a, l, p, g and w refer to maintenance, activity (walking), lactation, pregnancy, 17
growth and wool, respectively. 18
19
From GE, the methane emissions can be calculated from the GE intake using: 20
21
kg CH4/mature animal/year = gross energy intake (MJ/year) x 0.065/55.65, or 22
kg CH4/animal(<1 year-old)/year = gross energy intake (MJ/year) x 0.045/55.65 23
24
where the values of 0.065 and 0.045 refer to the 6.5 percent and 4.5 percent loss factors for methane 25
of gross energy intake, and 55.65 is the energy content of methane in MJ/kg. 26
27
Typical ranges for values of DE % are: concentrates: (75-85 percent), pasture (65-75 percent) and 28
low-quality forage (45-65 percent). 29
105
In practice, DE % will vary during the year and an example of this from the NZ GHG Inventory for 1
average sheep pasture in New Zealand in winter, spring, summer and autumn at 73.8 percent, 2
77.7 percent, 68.1 percent and 66.1 percent, respectively (Pickering, 2011). Corresponding ME 3
concentrations are 10.8 percent, 11.4 percent, 9.9 percent and 9.6 MJ ME/kg DM, respectively. 4
5
http://www.fao.org/partnerships/leap